Over the last few weeks, I’ve talked about what remains human in Procurement at the individual role and subtask level, introducing the Human Edge Matrix, which allows practitioners to assess how much of what they do remains human, what is augmented, and what can be fully automated.

I also put forth my argument that what actually remains human - independent of capability - really isn’t even procurement’s decision to make. It’s shaped by what its internal customers and suppliers value and trust. As part of this argument, I laid out the seven outcomes that stakeholders actually care about and how to assess what remains human in the context of those outcomes.

All of this gives us a view of what stays human.

But it also raises the next obvious question: if we know what stays human, what do we actually build around it?

In other words, once we’ve identified the work that remains ours, we can’t just cobble together the remnants into our prior architectures. We need to rethink what the enterprise genuinely values - ideally, into something that cannot be eroded by the next wave of AI tools or absorbed by an adjacent function because ‘Procurement isn’t delivering commensurate value’.

To me, this is where the conversation needs to go next, especially for CPOs. Because if you’re sitting at your desk today, trying to build a Procurement function that will still matter in five years, you need a clear answer to the question: what am I building toward?

This post is my attempt at that answer.

A Different Way To Think About The Function

So what does the architecture of the Procurement organization look like in a Post-AI world?

While it’s tempting to think of this in the form of boxes on org charts or specific skills that need to be retained or developed, I think it’s more appropriate to visualize this architecture in the form of organizational moats. That is, capabilities that Procurement needs to build deliberately if it wants to remain relevant, valued, and credible in a post-AI world.

Note that I’m using the term ‘moats’ in the loose sense - borrowing from the competitive strategy world, where it refers to a structural source of defensibility (something that’s hard to replicate and that compounds over time). Michael Porter alluded to this through his Five Forces work, Warren Buffett popularized the idea in his shareholder letters, and Hamilton Helmer formalized it through his seven “powers” framework.

I’m not trying to shoehorn Procurement into any of those perspectives but, instead, I’ve taken inspiration from their models, especially the idea that defensibility should constitute both a benefit and a barrier, and applied it to what I believe a post-AI procurement function needs.

A quick note before I get into the moats: this is not a list of skills. Skills live inside individuals and I’ll tackle those in upcoming posts. For example, judgement is a differentiating skill but it’s embedded in all six of the moats I’ll describe. It’s not a moat in and of itself.

Similarly, the ability to develop the next generation of talent is also not a standalone moat, in my view. It’s the maintenance layer that keeps it all intact over time. Important but not distinct items on this list.

The moats I’ll discuss are the organizational capabilities that CPOs need to be deliberately building: defensible terrain that comprises the architecture of the function, not the people inside it.

The Six Moats That Matter

There are six moats I’ve identified:

• Two customer-facing moats (commercial partnership and execution orchestration)

• One internal-facing moat (hybrid operating model design)

• One supplier-facing moat (supplier intelligence broker)

• One structural moat (commercial accountability and governance authority) and

• One about perception (enterprise positioning).

These cover, in my mind, the major surfaces where procurement creates and defends value.

Some of these moats exist in mature form in some functions today, though most don’t. All of them, I’d argue, are the things a CPO should be investing in now if they want their function to matter in a decade.

Let’s dive into each one.

1. Commercial Partnership

The first moat is procurement’s ability to act as a genuine commercial partner to the businesses it serves - not a process gatekeeper, not an order taker and not the function that shows up late to tell someone their preferred supplier isn’t on the approved list.

I mean a genuine commercial partner - someone who understands the business well enough to help shape what it’s trying to achieve, knows the category well enough to translate between commercial reality and operational need, and has earned the right to be in the room when real decisions get made.

There are three components to this that have to be built together:

• Business intimacy - the ability to hold a substantive conversation with the ‘consumer’ of the category about strategy, market dynamics, competitive pressures, and operating model choices, without needing a translator

• Category depth - genuine expertise in the supply markets, cost drivers, and supplier landscape relevant to that internal customer’s work, so procurement can contribute ideas rather than just react to requests

• Stakeholder relationships - the accumulated trust and rapport that makes internal customers want to involve procurement early rather than late (or worse, not at all)

Why this is a moat: The combination of these three things is rare, takes years to build, and is nearly impossible for AI to replicate - because it depends on context, relationships, and tacit knowledge that only comes from being in the conversations. An AI tool can surface market data but it can’t sit in a room and read the organizational or political dynamics of a leadership team debating whether to restructure a category. Once Procurement has earned partnership status, though, that status is sticky - once accumulated, it cannot be rapidly built by internal ‘competitors’.

What CPOs should be building: A deliberate model for developing commercial partners, not just category managers. That means rethinking hiring (possibly business generalists with curiosity, not just sourcing specialists), broadening training paths (focused on business acumen and stakeholder management), redesigning development opportunities (rotations into the business, not just within procurement), and creating explicit time and permission for the relationship work (including embedding these outcomes into individual metrics).

2. Execution Orchestration

The second moat is procurement’s ability to actually make things happen - to take a commercial problem and mobilize the people, process, suppliers, and systems needed to deliver the outcome.

And yes, I know the procuretech crowd uses “orchestration” in a narrower sense these days, usually around orchestrating AI agents across a workflow. That’s not what I mean. I mean orchestrating the enterprise: turning strategy into delivered value across stakeholders, suppliers, internal teams, and whatever AI-assisted workstreams sit in the middle. (A more sophisticated form of project management, if you will.)

This capability spans two ends of the spectrum:

• Bespoke work - taking on a complex, ambiguous commercial problem that no playbook covers and figuring out how to deliver

• Routine work - ensuring that standard buys happen without friction so the business never has to think about them

Both matter, and they reinforce each other. The trust earned from frictionless routine work is what buys the credibility to take on the complex, strategic problems (though not on its own and not without Moat #1 above).

Why this is a moat: Orchestration at this level depends on influence without authority, which is one of the hardest organizational skills to build and one AI has virtually no ability to replicate. It also compounds. A function known for making things happen gets invited into more things, which creates more opportunities to demonstrate the capability, which deepens the reputation further. Procurement’s very own flywheel.

What CPOs should be building: Explicit development of this type of orchestration muscle - which means giving people increasingly complex delivery challenges with real accountability, resisting the temptation to reduce orchestration to a process manual, and coaching them through the political and influence dimensions. Orchestration in this sense relates to judgment, intelligence, flexibility and an outcomes orientation - not a process orientation. You need a particular caliber of individual to be able to do this.

3. Hybrid Operating Model Design

The third moat is the internal architecture of how procurement work actually gets done (and the closest I’ll get to the idea of boxes on org charts in this conversation).

This involves the deliberate design of the Procurement operating model so that it’s clear which decisions route to humans, which route to AI, where the handoffs sit, how exceptions get escalated, how the AI tools themselves get governed, and how the function’s own workflow is structured so that humans stay sharp rather than getting progressively deskilled (or distracted by low value activities).

I appreciate that this can sound like an overlap with the first two moats but it’s actually quite different in nature. Commercial partnership is about how procurement engages the enterprise. Orchestration is about how it delivers outcomes. The operating model is about how procurement organizes itself to sustain both over time in a hybrid human-AI environment.

It’s effectively the plumbing, which most functions haven’t deliberately built as yet. The default approach to AI in procurement today is additive: bolt an AI tool onto an existing process, hope for productivity gains, and move on. That approach creates hidden liabilities:

• It optimizes locally without asking whether the overall flow still makes sense

• It embeds AI decisions in places where accountability may still be unclear

• It deskills the humans who used to do the work

A well-designed operating model does the opposite. It asks explicitly: where does human judgment add value here, and how do we protect the conditions that let it develop? Where does AI genuinely help, and how do we govern it? These are design questions, not technology questions - and answering them well is a capability in its own right.

Why this is a moat: Most procurement functions aren’t even framing these questions yet, let alone answering them deliberately. The ones that do will have a structural advantage in sustaining all the other moats because their humans will stay sharp and their work will stay coherent. As such, this is a meta-moat: one that holds the system together and allows the other moats to flourish.

(It’s also worth emphasizing that operating model design in a hybrid human-AI environment is an emerging discipline with very few practitioners. Building the internal capability to do it well is itself a cornered resource, because the people who can actually design these systems are few and far between, and will likely remain rare for years.)

What CPOs should be building: An explicit operating model design function - probably a small team or a named role - responsible for thinking about the human-AI-process architecture as a living system, not just implementing whatever the latest vendor sold them. This is among the least developed of the six moats in most procurement functions today, and the one where deliberate investment pays off fastest.

4. Supplier Intelligence Broker

The fourth moat is building on Procurement’s privileged position in the supplier ecosystem - specifically, the ability to extract genuine strategic intelligence from suppliers and translate it into value for the enterprise. This is one of the most underused sources of defensibility in the function today and, in my view, one of the most durable.

Here’s the core idea:

Suppliers - especially in strategic categories - know things the enterprise doesn’t. They see the market differently. They have perspectives on competitors, on technology trends, on regulatory shifts, on what other customers are doing. A procurement function with deep, trust-based supplier relationships can become an intelligence broker between that external knowledge and the internal decision-makers who need it.

But - and this is the catch - it only works if the relationships are genuinely trust-based. Which means procurement has to earn the right to that intelligence through years of reciprocity, discretion, and actually treating suppliers as partners rather than counterparties to be squeezed.

But most procurement functions don’t operate this way. The ones that do have built something extraordinarily hard to replicate, because the trust that enables the intelligence flow was accumulated over time and cannot be bought, copied, or AI-generated.

Why this is a moat: It’s a cornered resource in the strictest sense. The relationships are unique, the trust is non-transferable, and the intelligence flow depends on a position that only Procurement is structurally positioned to hold. AI tools can aggregate public supplier data; they can’t replicate a twenty-year relationship with a supplier’s CEO who is willing to share something they haven’t told anyone else. And the value to the enterprise - genuinely differentiated insight into the supply markets that matter most - is exactly the kind of thing internal customers will pay a premium for in attention, engagement, and budget.

What CPOs should be building: A deliberate architecture for supplier intelligence. That means identifying the categories where intelligence matters most, investing in the relationships that unlock it, creating internal mechanisms to translate supplier insight into decision-ready intelligence for business unit leaders, and - most critically - protecting the trust that makes the whole thing work by resisting the temptation to exploit suppliers for short-term cost gains. This moat is the easiest to destroy and the hardest to rebuild.

5. Commercial Accountability And Governance Authority

The fifth moat is different in character from the others. It’s not a capability procurement builds alone but a structural position procurement claims, formalized at the executive level, that establishes the function as the organization’s accountable owner for commercial and supplier decisions.

The question this moat answers is as follows: In a world where AI is increasingly recommending suppliers, flagging risks, scoring bids, and triggering contract actions, who is accountable when those calls turn out to be wrong?

The answer shapes where authority, budget, and relevance concentrate in the post-AI enterprise. And there are really only three possible homes for this accountability:

1. The business units that use the suppliers - which creates a clear conflict of interest, since the people making operational decisions shouldn’t be the same people overseeing them

2. A new enterprise function - like governance or technology risk, which gradually absorbs procurement’s commercial oversight role

3. Procurement itself - as the function with the cross-enterprise view, the supplier relationships, and the commercial context to bear it credibly

The third option is the one that preserves Procurement’s structural relevance. But it doesn’t happen by default; it has to be claimed, negotiated, and formalized - typically in partnership with the CFO and General Counsel, and ratified at the executive level. Once claimed, it becomes one of the most defensible moats on this list, because no other function is structurally positioned to take it over.

Why this is a moat: It’s a claim on organizational territory that, once established, is genuinely hard to dislodge. It also has an interesting property the other moats don’t: the threat isn’t primarily from AI itself, but from other functions - risk, legal, technology - that might otherwise absorb this accountability by default. Procurement that doesn’t claim this role loses it, and losing it erodes the rationale for procurement as a central function at all (which as I explained above, comes at a cost).

What CPOs should be building: The arguments, the relationships, and the internal capability to bear this accountability credibly.

• Step one is a conversation with the CEO and CFO about where commercial accountability for decisions relating to the supply base - human and AI-enabled alike - should sit (and why procurement is the right home).

• Step two is the internal capability - governance frameworks, decision audit trails, escalation protocols - that lets procurement actually discharge the responsibility once it’s been claimed

Without step two, claiming step one is reckless. Without step one, building step two is pointless.

This is the moat most dependent on executive alignment, and the hardest to build through sheer functional competence alone. Unlike the other five, you can’t start building this one tomorrow morning unless you already have the executive relationships to do so.

6. Enterprise Positioning

The sixth moat is the one Procurement has historically been worst at - and the one that most urgently needs to change.

It’s the deliberate work of shaping how Procurement is seen, engaged, and valued by the rest of the enterprise, including the service model that delivers on the positioning and the ongoing communication that sustains it.

Call it “branding” if you like, though I appreciate that that word triggers all the wrong associations in a Procurement audience. This isn’t about marketing fluff, logos, or internal newsletters. It’s about the gap between what procurement actually does and how the enterprise perceives what procurement does - a gap that, in most organizations today, is enormous and damaging.

If the enterprise sees procurement as a process gatekeeper, a cost cop, or a source of friction, then in a world where AI makes much of the process and friction disappear, the enterprise will stop routing work through procurement at all. The function gets disintermediated - for reasons not at all related to capability. Perception and reality have to move together, and right now they’re badly misaligned in most organizations.

Closing that gap is itself a capability, and it has to be built deliberately because it won’t happen on its own.

What does good positioning actually look like?

• A clear, honest articulation of what procurement offers that no one else in the enterprise can

• A service model designed around outcomes rather than process - for example, an account executive approach, where named procurement leads own the relationship with specific business units and are measured on commercial outcomes rather than activity metrics

• Sustained, proactive communication that keeps the function’s value visible rather than assuming it’ll speak for itself

Most procurement functions do none of these well. The ones that do, stand out immediately.

Why this is a moat: Positioning, once established, compounds the same way brand does in consumer markets. A Procurement function that’s understood as a commercial partner gets invited into more strategic conversations, which produces more opportunities to demonstrate value, which reinforces the positioning. A function that’s understood as process overhead gets routed around, which reduces its visibility, which accelerates its decline. The flywheel runs in both directions, and the direction you end up in depends on whether you invested in the positioning work when it mattered.

What CPOs should be building: A deliberate positioning strategy, treated with the seriousness a marketing function would treat its brand. That means defining what procurement stands for, designing the service model that delivers on that promise, and creating the ongoing rhythm of communication and engagement that keeps it alive in the enterprise’s mind.

(Of course, it also means honestly assessing the gap, as uncomfortable as it may be - between current perception and desired perception, and building a plan to close it.)

One brief acknowledgment on this moat: it’s possible that in ten years, positioning won’t need to be a distinct capability, because Procurement will have rebuilt itself enough that the function’s value is self-evident. I don’t actually think we’ll get there that quickly - but even if we did, it would be a good outcome. For now, though, the gap is real, and pretending it isn’t would let CPOs off the hook for work that desperately needs to happen.

Where This Leaves Us

Six moats, then. Two customer-facing, one internal-facing, one supplier-facing, one structural, and one about perception and reputation. Each of them hard to build. Each of them defensible once built. Each of them compounding over time.

Two final points:

First, it’s worth noting that these six moats are not independent of each other:

• While Commercial partnership and Execution Orchestration can exist without the Hybrid Operating Model Design moat, they won’t sustain in a post-AI environment unless the redesign happens

• Commercial partnership and Supplier Intelligence Broker reinforce each other, as the insights from suppliers are often what make procurement valuable to internal customers in the first place.

• Commercial Accountability & Governance Authority becomes credible only when the Commercial Partnership, Execution Orchestration and Hybrid Operating Model Design moats are in place to credibly justify it.

• Enterprise Positioning will be hollow without the substance of all of the other moats before it.

A CPO thinking about where to start cannot, therefore, treat these as six separate investment decisions; they form a system, and the sequencing matters.

Second, if you’re a CPO reading this, ask yourself this: how many of these six do I have in any meaningful form today?

My own answer, based on the functions I’ve worked with across my years at A.T. Kearney and The Smart Cube, is that most have fragments of one or two, almost none have all six, and the most neglected moats aren’t always the ones you’d expect.

I want to come back to this topic in a future post, specifically on the topic of where a CPO serious about post-AI relevance should actually start.

For now, though, the framework itself is the starting point.

The CPOs who invest in these six moats now will run the procurement functions of 2030. The ones who don’t will be running something smaller, more marginal or possibly nothing at all.

If you read last week’s post, you’re left with an uncomfortable question:

“If what stays human in procurement is determined by stakeholders rather than by the function itself, how does a practitioner actually figure out where the line falls?”

It’s one thing to accept that your relevance is shaped by the people you serve. It’s another to know what to do about it.

This week, I want to give you a tool for answering that question: For each key outcome, how do we determine what stays human and what doesn’t?

Why The Matrix Isn’t Enough

Of course, one approach could be to apply the Human Edge Matrix to each outcome, which I discussed in this post a couple of weeks ago (along with an interactive tool that you can use to assess how vulnerable your own role is to AI).

However, the Matrix was designed to classify work at the role and task level; it tells you whether a specific activity should be automated, augmented, or kept human.

For example, I’d posit that outcome #1 (Speed and Responsiveness of the Procurement Process) can be almost entirely machine in its execution (perhaps with some/limited human involvement where needed) while outcome #8 (Crisis Management) is one that will almost certainly remain fully human (even if those humans are somewhat augmented with AI.

But, in between these extremes - depending on the individual company situation (influenced by everything from its market, competition, financials, category focus, etc.) - all eight factors in the matrix could run the gamut from human to nonhuman.

What we need now is a complementary lens that operates at the outcome level and answers the question: “For this outcome, does human involvement change what the stakeholder receives?”

Two Questions That Draw The Line

So what is the right way to understand what stays human at the outcome level?

At least at the Internal Customer level, I’d suggest two core questions need to be answered:

1. Does human involvement produce a value premium that justifies the cost AKA “Am I making this better”?

   1. This question is purely economic: is the outcome measurably better, or perceived as meaningfully more legitimate, when a human is involved? And is that difference worth what the human costs?

   2. This covers a host of considerations including:

      1. Relative quality differentials i.e. is there a material quality differential between the ‘human only’ versus ‘human plus machine’ versus ‘machine only’

      2. Business partner requirement - does the work require Procurement to partner with the stakeholder to arrive at an optimal solution? Does he/she bring advisory value to the table?

      3. Experience levels - Does the practitioner bring a depth of experience and insight that makes a difference?

2. Does accountability require a human AKA “Does someone need to own this”?

   1. This question isn’t about whether AI can “do the analysis” but “can the organization accept a decision where no human bore the responsibility?”

   2. This covers a host of issues including:

      1. Regulatory sign-off

      2. Ethical and/or value-based oversight

      3. Complexity that demands multiple human eyes (for validation or risk mitigation) and,

      4. Situations where someone needs to be personally answerable for the outcome.

This allows us to create a simple 2x2 that connects naturally to the Human Edge Matrix as a complementary lens rather than a competing one: Accountability required / not required on one axis, value premium present / not present on the other - giving us four quadrants, each with a clear implication for the internal customer:

• Accountability Required:

  • Human Value > Machine Value: Procurement should handle

  • Human Value < Machine Value: Procurement needed ONLY if it can show material, incremental value: technical knowledge, regulatory understanding, etc. otherwise stakeholders stop calling (risk of rogue duplication)

• Accountability Not Required:

  • Human Value > Machine Value: Procurement should handle

  • Human Value < Machine Value: Automate and retain with Procurement IF there is consistency across customers, ELSE co locate with customer

The framework should be applied at two levels and at different cadences:

1. The Category Strategy level:

   1. When a CPO or category leader is designing or redesigning how a category operates, they apply the 2x2 to the outcomes that matter for that category.

   2. This is a periodic, strategic exercise. You do it when you’re setting up the category strategy, and you revisit it when something material changes e.g. new regulation, new technology capability, a shift in what the business expects from that category.

   3. The point is not to be recalibrating constantly, but factoring in this analysis at deliberate review points.

2. The Exception/Escalation level:

   1. In day-to-day operations, the default mode is whatever the category strategy determined.

   2. But specific situations will arise that challenge the default: a supplier relationship that was fine on autopilot suddenly needs human attention because of a quality failure or an internal stakeholder who was happy with automated reporting now needs human counsel because they’re facing a board question about supply risk.

   3. These are the moments where a practitioner applies judgment about whether the current situation has shifted the accountability or value-premium calculus. They use the two questions as a gut-check: has something changed about who needs to be accountable here, or about whether my involvement changes the outcome?

The Framework in Practice

Let’s apply this framework to three different scenarios:

Scenario 1:

Your VP of Manufacturing needs to consolidate your packaging supply base from five suppliers to two

This impacts the Total Cost of Ownership and the Supply Resilience outcomes.

The AI can do a lot here - spend analysis, supplier performance scoring, TCO modeling across the five suppliers, scenario modeling for different consolidation options and more. And it can do all of this faster and more comprehensively than any human analyst.

But let’s apply the two questions:

• Does accountability require a human?

  • Yes - consolidating from five to two suppliers is a decision that increases concentration risk. If one or both of the remaining two suppliers fail, production is materially impacted.

  • Someone needs to own that call, explain the rationale to the plant director, and be answerable when the board asks why the company is now dependent on two packaging providers instead of five.

  • No organization is going to accept “the algorithm recommended it” as an answer when a production line goes down.

• Does human involvement produce a value premium?

  • Yes - but not where we might expect. The analytical work (spend modeling, TCO calculations, performance benchmarking) is exactly the kind of structured cognitive work that AI does well and arguably better than humans.

  • The value premium shows up in the judgment calls the data can’t make: which two suppliers have the management quality and financial stability to handle twice the volume? Which ones will invest in our relationship if we double their share? How will the three suppliers we’re exiting react — will they become hostile in other categories where you still depend on them or will there be political/social implications?

  • Those are questions that require contextual insights (market, relationship and commercial) that no model currently possesses.

This lands in the top-left quadrant: accountability required, human value premium present. Procurement should own this. But notice that the analytical work within this outcome can and should be machine-augmented. What stays human is the judgment, the stakeholder conversation, and the accountability for the decision.

Scenario 2:

Your Chief Compliance Officer needs assurance that a new raw materials supplier in SE Asia meets the company’s labor standards and environmental commitments before the first PO is issued.

This impacts the Compliance and Ethical Assurance outcome.

AI can do a significant amount of the legwork - screening the supplier against sanctions lists, pulling public records, analyzing ESG ratings from third-party databases, even scanning news sources for red flags. In fact, the machine will almost certainly be more thorough and faster at this screening than a human would be.

But again, let’s run the two questions:

• Does accountability require a human?

  • Yes - this is one of the clearest cases. Regulatory frameworks increasingly require demonstrable human oversight of supply chain due diligence decisions.

  • Beyond the legal requirement, there’s a governance reality: if this supplier ends up on the front page for labor violations, someone in the organization needs to have signed off on the decision to onboard them.

  • “We ran the algorithm and it came back green” is not a defense that any general counsel will accept.

• Does human involvement produce a value premium?

  • Only partially - this is where it gets interesting. For the screening and data-gathering work, the machine is arguably better than the human. It’s more comprehensive, less prone to oversight, and doesn’t get fatigued reviewing supplier questionnaire number forty-seven.

  • But the human value premium is there - even if it is narrow but critical. It shows up in the interpretation of ambiguous signals (the supplier’s ESG score is acceptable but their audit history shows a pattern of just-in-time remediation before inspections), in the ethical judgment calls (the data is technically compliant but something doesn’t feel right), and in the conversation with the CCO where someone needs to say “I’ve looked at this and here’s my assessment”.

This lands on the top-right quadrant BUT somewhere towards the left: accountability is required, but the machine does most of the heavy lifting. Procurement is needed, but primarily for the sign-off, the judgment on edge cases, and the ability to stand behind the decision. If the procurement person is simply rubber-stamping what the AI screening tool produces without adding interpretive value, the function is at risk of being reduced to a compliance checkbox.

Scenario 3:

Your Head of R&D wants to explore whether any of your existing chemical suppliers could reformulate a key input to reduce costs and improve sustainability - but she doesn’t know which suppliers have the capability or the willingness.

This is the Supplier-Enabled Innovation outcome.

And it’s worth noting what AI can and can’t do here. AI can scan supplier capability databases, analyze patent filings, identify which suppliers have R&D facilities working on relevant chemistry, and even draft an initial outreach brief. All of which is useful groundwork.

So let’s, then, run the two questions:

• Does accountability require a human?

  • Not really - at least not in the regulatory or compliance sense.

  • Nobody is going to get fired or face legal consequences for how the innovation exploration was conducted. There’s no structural mandate for human sign-off on “let’s have a conversation with Supplier X about reformulation possibilities.”

  • This isn’t a risk or governance question.

• Does human involvement produce a value premium?

  • Overwhelmingly yes - this is perhaps the clearest case of human value premium across all seven outcomes. Innovation from suppliers doesn’t happen because you send them a brief.

  • It happens because a procurement professional who has built trust with the supplier’s technical team over years picks up the phone and says, “I think there might be something here. Can we get your head of applications science in a room with our R&D director?” It happens because the procurement person understands both sides well enough to see the connection that neither party would see on their own. It happens because the supplier’s commercial director is willing to invest internal resources in the exploration because she trusts the procurement person’s judgment that this company will actually follow through (and not just run a free innovation workshop and then give the business to a cheaper competitor).

No AI can replicate the relational capital, the cross-organizational pattern recognition, or the credibility that makes a supplier say “yes, we’ll invest our best people in this.” The machine can identify the opportunity. The human creates the willingness.

This lands in the bottom-left quadrant: no accountability requirement, but strong human value premium. Procurement should own this - but there is a nuance: it has to earn it. There’s no structural mandate keeping this work in procurement. If the R&D director doesn’t believe the procurement person adds value to her supplier innovation conversations, she’ll go directly to the suppliers herself. Procurement’s ownership of this outcome is justified entirely by demonstrated value, not by rules or policies. And that makes it both the most rewarding and the most fragile kind of human work.

(It’s worth dwelling a little on this last point: The work that has the highest human value premium but the lowest accountability requirement is the work practitioners most need to protect. Unless they are genuinely, demonstrably good at it, there will be no requirement to keep it with Procurement and/or work with the function.)

The Supplier’s Test Is Simpler - But No Less Important

You’ll notice I mentioned earlier that the 2x2 is primarily a tool for thinking about the internal customer relationship. So what about Procurement’s other prime stakeholder: the supplier?

Well, the supplier/market frame operates on a different logic.

When a supplier evaluates whether they need a human counterpart in Procurement, they’re not running an accountability-versus-value-premium calculation. They’re asking something more fundamental:

• Does this person have the authority to commit?

• Do they understand our business well enough that I don’t have to re-explain our constraints every quarter?

• Will they still be here next year, or will I be rebuilding this relationship from scratch with their replacement?

In other words, the supplier’s test for what stays human is about continuity, authority, and contextual depth. For the supplier, there is real signaling value in human engagement.

Because the fact is that a supplier will accept automated purchase orders, automated invoice processing, and even automated performance measurement.

But what they won’t accept - at least for relationships that are important to them - is a rotating cast of humans with no institutional memory, or worse, no human at all when they need to have a difficult conversation about pricing, capacity, or priorities.

This is, therefore, a simpler assessment than the internal customer 2x2, but it carries its own implication: the human work that matters most on the supplier side is relational infrastructure.

And relational infrastructure takes time to build, is easy to destroy, and impossible to automate.

Four Things The 2x2 Doesn’t Show You

First, decision making cannot be a cold process.

We are humans after all, and hence have to make reasonably human decisions. And so there is a rational case to be made that, in many instances, human involvement has economic value precisely because people aren’t rational about it. The empathy research discussed in the last post shows us this: the “human empathy premium” is an economic fact, not a sentimental plea.

Second, none of the above should be seen as an argument to reduce the practitioner to becoming a passive recipient of stakeholder judgment.

“What stays human is what your stakeholders demand to be human” can be seen as disempowering but that is not at all the point. The practitioner has the opportunity to shape stakeholder expectations rather than merely respond to them, because the best procurement professionals don’t just answer what stakeholders ask for, they influence what stakeholders think they need.

This means orchestration and judgment and creativity and outcome orientation beyond the traditional cost savings rubrics.

Third, what happens when the two stakeholder groups - internal customers and suppliers - disagree?

A business unit might be perfectly happy receiving an AI-generated market analysis and never speaking to a procurement person. But the supplier on the other end of that same category might need human engagement for the relationship to function e.g. contract renegotiations, performance conversations, innovation discussions, etc.

The reverse is also possible: a supplier might be fine dealing with automated PO systems while the internal stakeholder insists on a human procurement partner for strategic advice.

These situations will occur, and it’s worth acknowledging that the two frames can produce conflicting signals and that navigating that conflict is itself irreducibly human work.

Finally, we have to note the temporal factor. That is, things do and will change.

Whatever our analysis shows today (especially about AI’s current capability thresholds), its worth noting that that is the current view. It’s incumbent on us to keep reading the signals as the window moves.

The Real Point

The point of this framework is not to identify what’s ‘irreducibly’ human, because that line will keep moving as AI improves. The point is to identify what’s economically and organizationally irrational to hand to a machine, even if you technically could.

So it’s worth asking two questions: Does someone need to own this? Am I making this better?

If the answer to either is “yes”, the work stays human. If the answers to both are “no”, it doesn’t, regardless of tradition, comfort, or sentiment.

And if you’re sitting in a quadrant where accountability isn’t required and your value premium is the only thing keeping you relevant, you need to see that as a signal, not a safety net.

Because your premium has a shelf life. The question is what you’re doing to extend it.

In my last post, I introduced The Human Edge matrix, a framework for classifying procurement work in the AI age - what should be automated, what should be augmented, and what should remain human.

(I also included an interactive tool to apply the framework to your own role to assess how much of it should be done by AI as well as which aspects should remain human.)

The response was encouraging, but one question keeps surfacing in conversations with practitioners and leaders:

“Sure, but will any of this still be human in five years?”

It’s a fair question. And the honest answer is: probably not all of it.

The boundary between human and machine work is not a fixed line; it’s moving - but only in one direction. That doesn’t mean everything eventually will go to the machines. Some work will remain human for a long time, perhaps permanently. The question is: what work is that and who makes that call?

Here’s what I’ve come to believe: it’s not procurement’s decision to make.

In this and next week’s post, I’ll get into why this is as well as how to assess what Procurement work stays human and what goes the way of the machine.

The False Binary

Is there any work that is irreducibly human?

I’ve been thinking about this question a lot lately, as have many others. If you read the popular press, you can’t help but be pulled in two completely different directions.

At one end of the spectrum are the Dismissers, those who believe that AI is overblown and there is, and always will be, plenty of work that is and should remain human.

Some of this is based on beliefs that are grounded in sentimentality and emotion, but many dismissers have valid reasons to feel this way. AI hallucinates, it makes mistakes, it can’t do the simplest things (for humans) like read the room. It also lacks real world context, doesn’t understand people in the full human sense, or make thoughtful trade-offs based on variables not codified in the data.

But it’s also worth remembering that AI is still in its infancy, so we should expect the technology to continue to improve. It has already improved considerably in the last few years, and as this improvement continues, we should expect its capabilities viz-a-viz work traditionally done by humans to expand. So it’s entirely plausible that what is human today may not remain human in future.

That’s not to say all of it will go the way of the machine.

At the other end of the spectrum we have the Utopians, the believers who argue that AGI will be upon us soon enough, and that we will have amongst us a superintelligence that will be able to do everything a human can do and more. There will be little or nothing that won’t be done by machines, allowing us all more time to do all we ever aspired to do.

I get where they’re coming from as well - capability gains are arriving faster than ever, the machines are becoming multi-modal and agentic, and there are a ton of incentives, suggesting that human-like general intelligence may not be far off.

But today’s systems still confuse fluent performance with genuine understanding, and I believe that we don’t even know what we don’t know in terms of our brain’s architectures. As such, I personally don’t believe AGI is realistic or achievable, at least not in its full utopian form, any time soon.

What AI Can Fake…And What It Can’t

That said, there is some research that suggests AI has qualities that border on the “human” - and even improve on them in some ways.

For example, there is some evidence that AI can model ‘empathy’, though not without caveats. Specifically, research over the last couple of years finds that:

• AI is clearly getting good at empathy-shaped language - that is, for many text-based tasks, LLMs can produce responses that people rate as highly supportive, compassionate, and emotionally intelligent, even outperforming humans in benchmarked test.

• But this is mostly evidence of simulation, not sentience - they show patterns of empathy (cognitive empathy and emotionally appropriate language), with no evidence of real empathy (concern, emotion, consciousness, etc.)

• In addition, human authenticity still matters - even where AI responses are rated as excellent, people value empathy more when they believe it comes from a human.

• Labeling effects matter a lot - there is a ‘human-label’ premium; AI could outperform in building closeness when labeled as human but explicit AI-labeling reduced closeness.

(For the specific sources, look here, here, here, here, here, here, here and here.)

It’s Not About Procurement…

We can draw some interesting insights from this research.

Broadly speaking, people accept AI for analysis and understanding, but resist it for emotional sharing and genuine care. As good as AI might be, the value of human involvement in certain work is not about capability but about perceived legitimacy and trust (that is, perceived intention, shared vulnerability, and the belief that another mind is genuinely with you).

In other words, what people value isn’t capability but perceived human investment, which means that what stays human is going to be determined by what stakeholders value, not by what machines can or can’t do.

This leads to an uncomfortable conclusion for procurement professionals: the question of what stays human isn’t entirely for Procurement to answer. It belongs to the people it serves and the markets it manages; their willingness to trust, to accept, to engage, is what draws the line between human and machine work. Which means if we want to understand what remains human, we need to stop looking inward and start looking outward.

…It’s About The Stakeholders

Of course, there are multiple stakeholders, including the CFO, the board, regulators, internal customers, and suppliers. While the CFO and the board shape what Procurement is measured on, theones that shape how the work gets done are the two constituencies “external to the function”:

• The “Internal Customer” (Who Procurement Serves):

  • What do the internal customers that procurement serves (the business units, support functions such as marketing or HR or legal etc.) think about the work that procurement does?

  • What are the things that they absolutely want a human to do?

  • In which situations do they absolutely want a human to interact with?

  • What will they NOT trust from a machine output?

  • What do stakeholders actually pay for (in attention, trust, political capital) when they engage procurement?

• The “Supply Market” (Who Procurement Manages):

  • Given that Procurement represents the organization to and manages suppliers (and supply markets at large), how do they perceive the work that the function does?

  • In which instances do they require a human presence for reasons that go beyond capability?

Outcomes, Not Tasks

So what kind of Procurement work do these stakeholders value?

There are two ways to look at this - at the role/task/subtask level (the actual work Procurement does) or at the outcome level (the key outcomes expected by each stakeholder group)

The task/subtask levels is where most practitioners normally look to answer the question of what remains human. But there’s a problem with this approach.

Procurement’s stakeholders don’t care about the practitioner’s task list. A marketing director who needs a creative agency contracted doesn’t think about whether the procurement person ran a three-bid process or used an AI-assisted evaluation tool. They care only about whether they got a capable agency, on reasonable terms, without it taking months to complete. Therefore, the practitioner’s role/tasks is too insular a view to take. It doesn’t take into account the perspectives of those the function serves.

These outcomes, on the other hand, represent what Procurement’s prime stakeholders - those it serves and those it manages - actually care about. Not policies, not processes, not tools, but outcomes.

As such, outcomes are, then, the most appropriate basis upon which to conduct this analysis.

Seven Outcomes That Define Procurement’s Value

So, what are those outcomes?

Broadly speaking, there are seven broad outcomes worth considering:

1. Speed and Responsiveness of the Procurement Process

   1. When a business unit needs something purchased, how quickly and painlessly can they get it? The point here isn’t just speed but overall efficiency and effectiveness i.e. is the function seen to be an enabler or a bottleneck? Think elapsed time from request to delivery, in how many times they had to chase someone, and whether the process felt proportionate to what they were buying.

   2. A three-week sourcing exercise for a $5,000 software license is a failure of responsiveness regardless of how well the process was executed.

2. Achieving Optimal Total Cost of Ownership

   1. This goes beyond the purchase price to encompass the full economic cost of a buying decision, including implementation, maintenance, switching costs, quality failures, etc. From a stakeholder standpoint, this manifests itself when the cheapest option ends up costing more in rework, downtime, or internal frustration.

   2. The valued outcome here is not a 12% reduction of the unit price, but that the money spent yielded the highest utilization and produced the best possible return over the life of the relationship.

3. Maintaining Supply Resilience and Mitigating Risk

   1. This is a ‘sleeper’ outcome i.e. you only notice it when there is a disruption. Can the business count on having what it needs, when it needs it, without disruption?

   2. Is the category’s distinct risk profile (supplier dependence, geopolitical exposure, commodity volatility, technology obsolescence, regulatory change, cyber threats, etc.) understood and being well managed on their behalf?

   3. The valued outcome here is supply continuity that encompasses not just reliable delivery but supply base resilience itself. The goal is also not to eliminate risk but to ensure risks are identified, quantified if possible, and either mitigated or consciously accepted with the right people informed.

4. Ensuring Compliance and Ethical Assurance

   1. Stakeholders need to know that what they’re buying, and who they’re buying it from, won’t expose the organization to legal, regulatory, or reputational harm. This covers a host of factors, including sanctions screening, labor and environmental standards, anti-bribery obligations, data privacy requirements, and industry-specific regulations.

   2. The outcome isn’t just that the organization passed an audit, but that the business can operate confidently knowing procurement has built guardrails that protect them from risks they may not even be aware of.

5. Ensuring Optimal Supplier Relationships

   1. Suppliers are not interchangeable inputs; they each have their own capabilities, knowledge bases and priorities, and the quality of the relationship directly affects what the business gets from them.

   2. This outcome is about not just performance to agreed standards and remediating problems when they arise, but also whether the relationship is managed in a way that earns the organization preferential treatment (better allocation, priority access to new tech, faster response times, access to senior attention, etc.).

6. Driving Supplier-Enabled Innovation

   1. Some of the organization’s most valuable innovation doesn’t come from internal R&D but from suppliers who bring new materials, processes, technologies, or ideas that accelerate product development and time to market, improve quality, or open new market possibilities.

   2. This outcome is about whether procurement is positioned to unlock that value: identifying innovation opportunities, creating the commercial structures that incentivize suppliers to share their best thinking, connecting the right suppliers with the right internal teams, and ensuring the organization is seen by the supply market as a customer worth innovating for.

7. Crisis Management

   1. When something breaks e.g. a key supplier goes bankrupt, a pandemic disrupts global logistics, a geopolitical event closes a trade corridor, a quality failure triggers a recall, the organization needs procurement to respond with speed, judgment, and authority. This is episodic work that demands real-time decision-making in challenging circumstances.

   2. The outcome is that the organization survives the crisis with its financials, operations, relationships, and reputation as intact as possible - which calls for coordination across functions, direct engagement with suppliers and stakeholders, and the willingness to make consequential calls without the luxury of full analysis.

The Harder Question

The question that follows, then, is the harder one: for each of these outcomes, what must remain human, what can be augmented, and what should be fully automated?

That’s not a question you can answer in the abstract. It depends on who’s asking, what they’re willing to trust, and whether Procurement’s involvement makes the outcome measurably better.

In Part 2, I’ll introduce a framework for making that determination - one that’s grounded not in what AI can or can’t do today, but in what Procurement’s stakeholders will and won’t accept, and why that distinction matters more than capability ever will.

By now, you’ve probably seen the spider chart from Anthropic (below) that plots AI’s theoretical capability against observed AI coverage by occupational category. The blue area represents the share of job tasks that LLMs could theoretically perform; the red area shows the share actually being performed by AI, based on real world usage data from Claude.

The thing about this chart that had LinkedIn and other social platforms in a tizzy was not so much the red areas but the fact that the greatest theoretical exposure was in those occupations that perhaps even 5-10 years ago, we wouldn’t have thought susceptible to “automation” (I’m using that term in its broadest sense).

But that’s one of the prime implications of AI today.

Where AI Lands First

AI usually lands first on work that is repeatable, rules-driven, text/data-heavy, and separable from messy real-world context - work that just happens to be a lot of what the average LinkedIn user does: management, finance, IT, office administration, etc. (Yes, I appreciate there’s more to it than that, but that sort of work does form the basis of these functions.)

And as AI tools continue to improve (which they will), we can expect the gap between the blue and red dots to close.

None of this should come as any real surprise. There’s plenty of research and anecdotal evidence that this will be the case:

• The ILO’s 2025 exposure index assessed task-level exposure across nearly 30,000 occupational tasks and found that one in four jobs globally is exposed to GenAI to some degree, with clerical support roles still the most exposed. They highlighted that “some strongly digitized occupations have increased exposure, highlighting the expanding abilities of GenAI regarding specialized tasks in professional and technical roles”.

• KPMG cites that its own analysis indicates “50-80% of current procurement work can be automated, eliminated or shifted to self-service models”.

• More conservatively, McKinsey says that its analysis suggests that “technology will reshape the procurement function into an organization that is 25 to 40 percent more efficient, more agile, and increasingly agentic”.

Whatever numbers you choose to believe, it’s indisputable that AI, in one form or another, is going to change the way work is done.

For Procurement, the implications are obvious.

The earliest procurement impact is showing up in transactional and process-heavy cognitive work such as intake, data cleanup, PO support, contract reviews, supplier communications, first-pass analyses, and workflow orchestration.

But that’s just the start of it. AI is already coming for more Procurement work - including the analytical and decision support work that we previously thought would remain the domain of humans. We’re already seeing AI assist heavily with market intelligence synthesis, option generation, scenario modeling, contractual ‘red-flag’ detection, draft strategies, and negotiation preparation (though humans still own prioritization, trade-off selection, timing, and commitment).

The point is that machines are only going to get better - so the list of what AI can do will only keep expanding.

The Limits of a Task-Based View

The most fundamental takeaway for the practitioner, then, is that your role is going to change. There are (many) aspects of your role that AI will be able to do faster, cheaper and, yes, better (and not only that but it’s going to be able to do it 24/7).

But how exactly will your role be impacted?

There are plenty of institutions that have looked at specific Procurement roles and assessed the impact of AI on those jobs. Typically, they’ve taken a specific role, broken it down into its constituent tasks, and then assessed how susceptible each task is to AI.

In my view, this is useful but not enough.

Most procurement roles don’t fall into clean, well-defined sets of tasks. Practically, there are real-life complexities that force each role to morph in one way or another. These complexities can be external to the role (budget pressures, organizational or managerial demands, etc.) or specific to the individual (personal goals, expectations and desires).

As such, while these task-based analyses are helpful, the better question to ask is: how can we think differently about roles and really get to the root of what makes them human? This will allow individuals to determine for themselves why and how your particular role will be impacted by AI.

In this post, I’ll present one way to think about this: The Human Edge Matrix©.

A Better Way to Assess What Remains Human

The Human Edge Matrix provides us with a diagnostic structure, one that speaks to the nature of a given task or role and whether or not it will remain ‘human’ in the long term.

Specifically, there are two categories of analysis to consider with this matrix - the tiers of impact as well as the determining factors.

1. The Three Layers of Procurement Work

The first thing to understand is that this isn’t an “either/or” discussion. Every role won’t be either automated away or remain fully human. Work will split into three layers: Machine-Executable, Augmented and Human.

The three tier approach gives us a more realistic way to think about Procurement work.

It’s also worth noting that the assessment of what work falls within which tier is always going to be a point-in-time assessment. That is, while the 3 tiers hold, the work that falls under each tier is not static. As AI capabilities evolve, work currently in Tier 3 may migrate to Tier 2, and Tier 2 work may become Tier 1. It makes sense, therefore, to revisit any classifications periodically.

2. The Factors That Make Work More or Less Human

Within any role or set of tasks, a host of factors will determine where any given procurement activity falls in terms of the three tiers - eight to be precise.

Each of the factors operate as a spectrum, and it is the combination of factors, not just any single one, that determines classification.

These eight factors are as follows:

Factor 1: Codifiability

Can the decision logic, workflow, and success criteria be explicitly defined and systematized?

This encompasses both the structural clarity of the process (are there defined steps?) and the degree of precedent (has this been done many times before in similar ways?).

Highly codifiable work has clear inputs, known decision rules, and measurable outputs.

Example: Tail-spend PO processing against pre-approved catalogs is highly codifiable. Developing a category strategy for a new market with no prior supplier relationships is not.

Factor 2: Ambiguity

How much of the relevant context is tacit, situational, or absent from the available data? How rapidly is the relevant context shifting?

Ambiguity can be high for structural and dynamic reasons.

Structural ambiguity is high when the “right answer” depends on information that exists in people’s heads, in organizational culture, or in the dynamics of a specific moment. Hence, the the relevant context is tacit, relational, or simply not captured in available data.

Dynamic ambiguity is where the environment is changing so rapidly that the context for the decision is shifting faster than models or processes can incorporate it.

Example: A supplier’s public financials look strong, but the category manager has heard through industry contacts that the founder is planning to exit - tacit knowledge that could fundamentally change the sourcing decision. Allocation during supply crises, pricing shifts during geopolitical disruption or sudden regulatory changes create dynamic ambiguity (not because information is absent but because it’s changing in real time).

Factor 3: Judgment Complexity

Does the decision require weighing incommensurable trade-offs, interpreting incomplete signals, or making calls where reasonable people would disagree?

Note that this is distinct from ambiguity: a situation can be perfectly clear and still require sophisticated judgment. The question is whether the decision involves genuine dilemmas rather than optimization problems.

Example: Choosing between a lower-cost supplier with a questionable sustainability record and a more expensive supplier aligned with corporate ESG commitments. Both options are well-understood, but the judgment lies in how to weigh competing priorities.

Factor 4: Creativity

Is the work about optimizing within known parameters, or does it require imagining genuinely new approaches?

Optimization is AI’s strength. Genuine invention - new commercial models, unconventional partnerships, category strategies that redefine the problem - remains a human edge. The distinction is between finding the best answer within a known solution space versus redefining the solution space itself.

Example: Optimizing payment terms across a supplier portfolio is an optimization problem. Reimagining the procurement operating model to shift from transactional buying to outcome-based partnerships requires a creative rethink.

Factor 5: Stakeholder Complexity

How many stakeholders are involved, how conflicting are their interests, and how much does success depend on navigating those dynamics?

This encompasses both internal stakeholder management (business units, leadership, legal, finance) and external relationship management (suppliers, intermediaries, regulators). The underlying skills required - reading interests, building alignment, managing conflict - are the same.

Example: A routine MRO renewal involves one budget holder and one supplier. A strategic outsourcing decision involves C-suite sponsors, multiple business unit leaders with competing priorities, legal, HR, affected employees, incumbent suppliers, and potential new partners.

Factor 6: Political and Organizational Sensitivity

Is the work visible to senior leadership, does it touch on organizational power dynamics, or could it create reputational exposure?

Political sensitivity isn’t about the technical difficulty of the work but rather the organizational consequences of how the work/decision will be perceived. Identical analytical tasks carry different political weight depending on who is watching and what is at stake.

Example: Running a competitive tender for the CEO’s preferred consulting firm requires navigating political dynamics that have nothing to do with the mechanics of the RFP process itself.

Factor 7: Ethical and Values-Based Reasoning

Does the decision involve genuine ethical dimensions that require moral reasoning and alignment with organizational values?

This is less about compliance (which can be codified) and more about whether the organization’s identity and reputation are at stake. AI can flag such ethical risks, but the weighing of ethical trade-offs is fundamentally human.

Example: Deciding whether to continue sourcing from a region where labour practices are legal under local law but violate the company’s stated human rights commitments. No algorithm can resolve this as it requires a values-based judgment that the organization must own.

Factor 8: Decision Risk, Reversibility and Ownership

What is the magnitude of downside if the decision is wrong and can it be undone, and does the organization (or external stakeholder) require a human owner to stand behind it?

AI can handle high-volume decisions even if some are wrong, provided the errors are low-cost and correctable. Irreversible, high-stakes decisions demand human ownership. In addition, some decisions are auditable, require relationship legitimacy, and/or require an accountable human sponsor (even if AI did 80 percent of the work).

Example: Automatically reordering office supplies based on consumption patterns is low-risk and easily reversed. Signing a five-year sole-source contract for a critical component is high-risk and essentially irreversible. In other instances, a human will still be required to defend a decision to leadership, legal, the business, and/or a regulator.

How to Apply the Framework

Taking the three tiers and the eight determining factors together, the following matrix can be used as a diagnostic. For any procurement activity, assess where it falls on each factor. The majority of evidence will indicate its specific tier.

To apply The Human Edge Matrix to your own role, click the button below to access the interactive tool:

Interactive Tool

You’ll need to enter your name and email (you’ll be subscribed to my site) and then you can complete this assessment for your role at an overall level or by sub-task. Note that none of the information you input (other than your name and email) will be retained in any way. This is simply for your personal assessment.

(I’d love to get your feedback on the tool itself and whether you agree with its findings.)

What This Looks Like in Practice

The following examples show how specific procurement activities map against the framework. The tier assignment reflects the overall weight of evidence across all eight factors.

Tier 1 Examples: Machine-Executable

• Catalogue-based PO creation and approval routing for pre-negotiated items

• Invoice matching and exception flagging against contract terms

• Supplier onboarding document collection and compliance verification

• Automated spot-buy execution within pre-set parameters

Tier 2 Examples: Augmented

• Spend analytics and category spend classification

• Market intelligence synthesis for category strategy input

• RFP development and supplier response evaluation (AI drafts, human refines and decides)

• Contract redlining and risk identification (AI flags, human negotiates)

• Negotiation preparation: BATNA, scenario modeling, and playbook generation

• Supplier performance monitoring and scorecard generation with recommended actions

Tier 3 Examples: Human

• Category strategy development for volatile or strategically critical categories

• Cross-functional alignment on make-vs-buy, insource-vs-outsource decisions

• High-stakes, complex negotiations (multi-year, multi-party, novel deal structures)

• Strategic supplier relationship management and joint value creation

• Ethical sourcing decisions involving values trade-offs and reputational risk

Caveat: It’s worth noting that some procurement work will stop being human-executed before it stops being human-owned. That is, leaders may decide that there may well be work that remains human-supervised (even if AI can do it) because the task shapes learning and judgement.

From Task Taxonomy to Role Redesign

The goal of this framework is to provide a deeper way to think about AI’s impact on current roles, both overall as well as at the task level. It serves multiple audiences:

• CPOs and Procurement leaders: Use the matrix to audit your function’s activity portfolio. Identify which Tier 1 activities are still being done manually (automation opportunity), which Tier 2 activities lack AI tooling (augmentation opportunity), and which Tier 3 activities are being underinvested in because the team is trapped in lower-tier work.

• Procurement practitioners: Use the Tier-Factor matrix to assess your own role’s exposure to AI. The goal is to deliberately build capabilities in those areas that keep humans essential - judgment, stakeholder navigation, creative strategy, ethical reasoning, etc.

• For Procuretech leaders: Use the tier definitions to set realistic expectations for AI deployment. Tier 1 is ripe for full automation today. Tier 2 requires thoughtful human-machine workflow design. Tier 3 requires AI to serve as decision support, not decision maker.

One last point: What should emerge from this analysis is not just whether a role is at risk or to what extent - very few roles, if any, are going to survive intact in a Post-AI world.

What should emerge is a clearer indication of how to future-proof the practitioner for a post-AI world.

In addition, when you subtract tasks that will be automated - and even accounting for augmented tasks - what is left will almost certainly need to be rethought. The very nature of roles will need to be changed and, likely, rebundled across the function.

As such, Procurement roles will need to be redesigned around orchestration, exception management, business judgment, stakeholder alignment, supplier strategy, risk governance, and decision accountability, among other considerations. This will force us to move from a task taxonomy to a role redesign model. I’ll cover this topic more deeply in future posts.

In my last post, I flagged the idea of cognitive debt, which is the hidden cost we incur when a tool helps us complete a task without making us do enough of the thinking to truly understand, evaluate, or reproduce the outcome ourselves.

I also discussed how, if this debt goes unpaid, it leads to a loss of cognitive agency, which is the ability to understand the reasoning behind an outcome, assess whether it is sound, adapt it when needed, and take real ownership of the judgment (and output) involved.

This loss is a real problem, particularly in a world where we must grapple with information (and cognitive) overload, an attention economy, as well as performance metrics (explicit and implicit) that reward volume and activity over quality and individual development. The proliferation of AI tools is only exacerbating this problem, taking it to another level entirely. As I mentioned in my prior post, it’s the difference between creating sounds and developing musicianship.

And yet these tools aren’t going anywhere. Stand alone or embedded within more traditional SaaS tools, AI is becoming ubiquitous and will soon reach the point where it is no longer a feature but an expectation.

Used thoughtfully, they will dramatically enhance both our efficiency as well as our effectiveness - but only if we take the initiative. The onus remains on us to use them thoughtfully, not only for the betterment of our workplaces but for ourselves as well.

In this post, then, I’ll outline a practical framework for using AI in ways that preserve cognitive agency rather than erode it.

The Cognitive Agency Framework

At the heart of this framework is a simple principle:

Use AI to reduce mechanical effort, not to replace formative judgment.

But while this idea is simple in concept, it’s much more involved in execution. There is no point solution when it comes to the problem of minimizing cognitive debt and retaining our cognitive agency.

Our approach has to be multi-faceted - from how we lead to the parameters we set for it to what we individually must do. This is not a tool problem as much as it is an operating model problem.

And this operating model - The Cognitive Agency Framework, as I call it - has three layers:

1. Leadership (conditions)

2. Guardrails (workflow rules)

3. Individual Practices (specific habits)

Let’s look at each of these in turn.

Layer 1: Leadership Sets the Conditions

Everything starts with leadership, which must set the stage for their teams at the outset. The central question they need to address is:

“How do we use AI to capture efficiency gains without removing the cognitive reps that build and preserve judgment?”

There are four specific actions they should take:

1. Establish the Vision:

   Clarify the kind of practitioner the organization will value going forward. That is, enabled by AI tools, practitioners must:

   • Move from task completion to outcome ownership, where judgement and defensibility will become the differentiators

   • Remain focused on driving effectiveness as much as efficiency, which prioritizes achieving results for (internal) customers and moving the needle on their metrics.

2. Align incentives:

   If leaders only rewards efficiency metrics such as faster turnaround, more output or shorter cycle times, then they will end up (accidentally) training people to maximize AI throughput, not human discernment.

   Instead, they should balance these metrics (which will still be important) with more ‘human’ metrics that measure and reward on the basis of the progressive vision laid out in point 1. Specifically:

   • Measure for impact, not simply process compliance

   • Recognize innovation and progressive thinking in annual evaluation cycles

   • Reward reasoning quality and assumption strength, including risk awareness, contextual judgment, defensibility and trade-off understanding

   • Reward practical, sustainable AI deployment e.g. embedding AI into workflows

3. Build the Training Loop:

   Early, heavy AI reliance can weaken independent reasoning, ownership, and recall so it’s imperative to ensure that team members put in the mental reps needed to build their cognitive scaffolding.

   In other words, the key is to ensure that “judgment reps” are baked into the work itself.

   So set the expectation that each team member (in particular, junior team members) must have practical experience in “doing the work” (for specific types of work - see the next section on Guardrails) and not let AI remove ‘first-principle’ practices.

   This means, therefore, deliberately creating “unassisted reps” - for example, when it comes to issue framing or contract risk spotting, and even recommendation writing.

   For junior team members, this could be having them alternate between a solo first pass, an AI-assisted revision and then a (senior) human review.

4. Broadcast Learning:

   Every team will have individuals who are leading the charge when it comes to innovation and AI deployment. It’s important to identify and encourage these early ‘champions’.

   But long-term value for the function will only come from scale, and that starts with strong communications and information sharing. The more people find out about the potential, the early wins/realized value and the personal upside, the higher the chances of success.

   It’s important, then, to communicate broadly, which means:

   • Hold regular town halls to share AI wins, relevant use cases and key lessons learned

   • Invite demos by Builders to show the potential and value of current and emerging tools

   • Host hackathons on critical functional issues

   • Provide peer recognition, including symbolic as well as cash prizes

Layer 2: Guardrails Shape the Workflow

With the leadership posture defined, it’s important to set appropriate guardrails that help ensure intent turns into behavior.

This requires defining the specific rules that need to be put in place to ensure team members work with AI tools in the best, most optimal way. Five key guardrails should be stipulated:

1. Classify Tasks:

   In terms of allowing AI to run autonomously, what kind of work is OK versus what is not OK?

   Not all tasks should be treated the same; some need little to no attention while others need a human to keep driving it. As such, separating autonomous versus assistive versus formative work is an essential first step.

   Some tasks - such as formatting, information organization, summarizing, first-pass drafts, etc., are good candidates for heavier AI use.

   Others are formative in nature because they build or preserve judgment. Think:

   • Framing the problem

   • Identifying what’s missing in an analysis

   • Making trade-offs in difficult situations

   • Deciding what matters in moments of ambiguity

   • Defending recommendations

   These are tasks where humans need to stay in the loop. (No one is going to ask the algorithm for the justification of the decision. They’ll ask the human who put it forward.)

   As a practical rule, then, create a team-level list of:

   • Tasks where AI can lead

   • Tasks where humans must lead

   • Tasks where AI can support but not substitute

   Document this in a one-pager by workflow, potentially in an “AI RACI” format. This document should be reviewed at regular intervals (quarterly or semi-annual at minimum) as models and tools evolve.

2. Ensure No AI-First for Judgement Tasks:

   For specific tasks that require human judgement and involvement, stipulate that a human must always do the first pass. AI cannot be allowed to initiate the ‘thinking’.

   Embed the idea that people must think before they prompt, fleshing out their ideas, the core problems, constraints, etc. This brief doesn’t need to be a thesis, it can be relatively brief. But the discipline must be demanded; there needs to be a reasonable level of ideation and thought provided by the human as the first step.

   Post that first step, AI can then be utilized as the next step of assessment and input.

3. Use AI as Challenger:

   Never use the AI as a substitute or as the sole author of the output. Instead, treat is as a sparring partner. Reiterate that AI should be used to:

   • Expand options

   • Stress-test thinking

   • Surface blind spots

   • Improve articulation

4. Embed “Explain-Back”:

   Let team members know that they will be expected to explain the bases and implications of their analyses. Let them know that they will be asked whether they are comfortable owning the outcomes of their assessment and why.

   • What is the decision?

   • What trade-off is being made?

   • What would make this wrong?

   • What information is missing?

   • What is their level of confidence and why?

   If someone cannot defend their AI-assisted output, they cannot own it.

5. Insert Friction Gates:

   Well-placed friction preserves thinking quality.

   For high stakes categories and key process points that matter, clarify that team members will be asked to present key findings and rationalize their thinking.

   These key insertion points should be defined, and could encompass:

   • Award decisions over $X

   • Contract deviations that shift liability/indemnity/termination

   • Negotiation postures and walk-away thresholds with strategic suppliers

   • Risk acceptance (cyber, continuity, regulatory)

   • External communications that could create reputational exposure

   For particularly critical decisions, require red-teams be involved to challenge the analysis (e.g. “make the case against this” and ask the team to defend their point of view).

   The key here is simple: to raise the “cost of cognitive offloading” - when this cost rises, people offload less, retain and own more.

Layer 3: Individuals Build the Habits

Of course, the rubber meets the road with the individual. While the guardrails define the rules, individual practices determine whether a team member actually builds judgement.

As such, each individual’s goal must be to master ‘Attention Sovereignty’, that is, to actively direct attention rather than surrendering it to the algorithm. Attention sovereignty is the precondition for everything below.

Four key habits are essential here:

1. Develop Your Point of View First:

   One of the biggest protections against cognitive debt is requiring human pre-processing before AI enters the picture. So ask people to think before they prompt.

   For example, before using AI, require the user to first write:

   • the problem statement

   • the desired outcome

   • the likely risks

   • their own first-pass recommendation

   Even a brief set of notes or hypothesis or key framing questions and ideas helps. Start with a human frame first.

2. Use AI as Sparring Partner:

   With the human frame fleshed out (even at a high level), use AI as a challenger, expander, and/or editor.

   The safest pattern to deploy is not “do it for me” but to iterate with you. Ask it to work with you as a consultant or analyst. Ask it to:

   • Challenge your assumptions

   • Identify three risks you may be missing

   • Critique your recommendations

   • Provide you with alternative considerations

   • Test your logic for holes

   • Help you compare different scenarios

   At the same time, thoughtfully evaluate what it gives you back, and test its thinking to ensure that what it’s telling you is something you agree with. Ask it questions, pressure-test key statements and ideas, push back where you feel push-back is needed. Take nothing for granted.

   This preserves human ownership of the core judgment while still harvesting the benefits of the tool’s speed and breadth.

3. Build explain-back discipline by using a checklist

   This is a simple safeguard but also a powerful one. For every key analysis that leverages AI heavily, be able to explain:

   • What the recommendation is

   • Why it makes sense

   • What assumptions it depends on

   • What the risks are and where could it fail

   • What you changed from the AI output

   The practical rule to follow is:

   “I will not submit an AI-assisted recommendation until I can fully defend it my own words.”

4. Engage in Regular Self-Critiques:

   Regularly reflect on your thought processes, biases, and how you learn. Consider how you can continue to push your thinking and ownership of your work and analyses.

   Make a note of the various tools you utilize and their relative strengths and drawbacks. Establish personal “red-flag” triggers about when to use AI and when not to, based on individual usage.

   Adopt a two-source rule for high-stakes facts. Treat AI as a draft, not a source. Be selective about facts provided by AI tools and verify critical claims against primary documents or an independent reference.

   Reconstruct from memory. After using AI, restate the reasoning without looking. If you can’t explain it cleanly, you’ve borrowed output without building understanding.

The Procurement Cheat Sheet For Leaders

The three layer Cognitive Agency Framework is a useful way to think about retaining cognitive agency, both as a leader and an individual practitioner. But it does require a good measure of work to ensure it’s implemented effectively in any organization.

While in my view, the realized value is worth that work, I appreciate that ‘speed to action’ is more important than ‘perfect implementation’. To that end, if you do nothing else then, at minimum, implement the following four rules of thumb:

Rule 1: Define “When AI Should Be Avoided”

Some contexts are fragile and the downside of a subtle error can be asymmetric, in that small mistakes can create outsized legal, financial and/or reputational consequences.

As such, heavy AI reliance is best avoided in specific defined contexts. These could include:

• Sensitive negotiations

• Legal commitments without counsel review

• Reputational risk communications

• Compliance/regulatory issues

• Situations requiring confidential data handling policies

Rule 2: Preserve Human-First Reps in Judgment-Heavy Workflows

Make clear that Humans must think first and foremost when it comes to Judgement-heavy workflows, which you should spend a little bit of time thinking through. These workflows could include:

• Supplier selection/award situations

• Risk acceptance decisions

• Negotiation strategy and walk-away scenarios

• Contract deviations and redlines that shift risk

• Stakeholder trade-offs and prioritizations

• Financial/business case assumptions

Rule 3: Let AI Challenge and Draft, Not Decide

Have your team develop the first draft of any analysis or output. AI can then assist by:

• Summarizing bids

• Drafting supplier emails

• Structuring comparison tables

• Synthesizing documents

That said, the human should still own:

• Supplier selection logic

• Trade-off decisions

• Stakeholder balancing

• Risk acceptance

• Negotiation posture

Rule 4: Require Defense, Not Just Delivery

A recommendation must not be considered complete until the owner can explain the:

• Logic

• Risks

• Alternatives rejected

• Contextual factors

This ensures the organization is developing professionals, not simply ‘output assemblers’.

Closing Thought

AI is going to keep getting better. The real question is whether we (as leaders and as individual practitioners) will get better with it.

If leaders reward throughput, teams will optimize for throughput. If workflows don’t require defendable reasoning, people will stop building it.

Cognitive agency won’t survive on its own or by some accident. It will survive by design: when our systems make thinking both mandatory and unavoidable.

It’s incumbent then on leaders to set the conditions and establish the guardrails that shape the workflow, and for individuals to practice those habits that keep their judgment sharp.

This is the aim of the Cognitive Agency Framework. Use AI to remove mechanical effort, but protect the cognitive reps that build discernment. Otherwise, we’ll ship polished outputs but, over time, lose the capability underneath them.

The ultimate goal is simple: capture speed without surrendering musicianship.

AI is an incredible technology - one that’s taken the promise of classic SaaS and amplified it exponentially.

We now have tools that can help us produce better-looking work, faster, and with less friction than ever before. That’s a real and massive gain, but it’s one that comes with a hidden risk: these very same tools make it easier to skip the mental work we need to be doing to truly own our work. Or, to be more precise:

What happens when the tools that improve our outputs also reduce the amount of formative and evaluative thinking we do to truly understand, evaluate, and own them?

Creating Sounds vs. Developing Musicianship

Let me explain this with an analogy from the music world.

A couple of decades ago, if you wanted to record an album, you had to learn to play an instrument, put together a band, hire out a studio, bring in a producer and an engineer, record multiple takes and then piece together and master the final output.

Today, you can bypass much of that. Modern technology has made it easier than ever to create something that sounds polished. Synths, virtual instruments, multitrack recording, editing software, and now a host of AI tools that do all of the above with just a prompt, help generate impressive results in a fraction of the time.

And yet, there remains a difference between creating sounds and developing musicianship.

Musicianship means developing a sense of taste, timing and feel. It means having an ear for what works, understanding when to elevate tension and when to release. It means developing an understanding of which elements work together and why.

Developing musicianship requires doing the work. But creating sounds? Today, anyone can produce something that sounds good without developing the underlying fluency needed to create, diagnose, adapt and finalize with intention.

This same distinction is playing out in other spheres as well, including Procurement knowledge work. And that presents us with a challenge.

Cognitive Debt and the Erosion of Agency

In Procurement, we also have access to a host of incredibly powerful AI tools, but their deployment is all across the map.

I don’t just mean their usage but the intentionality of that usage. Some are using them well, others less so (often because these tools sound so authoritative and confident that users mistake their polished language for sound judgement).

This undisciplined and unintentional deployment in our knowledge work comes with a cost. It creates a cumulative risk: repeated cognitive outsourcing creates Cognitive Debt which, over time, erodes our Cognitive Agency.

Let me explain both of these terms briefly.

Cognitive debt is the hidden cost we incur when a tool helps us complete a task without making us do enough of the thinking to truly understand, evaluate, or reproduce the outcome ourselves. It’s a real problem because the convenience of today’s tools can, if we’re not careful, result in a complacency of understanding. That is, convenience borrows against comprehension, with very real, very material trade-offs.

The result, if this debt continues to go unpaid, is a lack of cognitive agency, which is the ability to understand the reasoning behind an outcome, assess whether it is sound, adapt it when necessary, and take real ownership of the judgment involved and, hence, the output generated.

This ‘cognitive offloading’ has real and practical implications: these tools might improve immediate task performance, but they do so while also reducing retention and internal encoding (i.e. weakening the depth of neural processing required for learning and recall), especially when our goal is to ‘grow’ as much as it is to just ‘get the work done’.

Why This Matters in Procurement

Now, this might sound like a problem for just the new entrants into the function, but it’s really a problem for all practitioners, junior and senior alike.

For juniors, the risk is failing to build the foundation of what makes for true, high quality performance in the long run. No ‘cognitive scaffolding’ being built in the first place, no formation of those foundational thinking skills that are so essential to ‘good judgement’.

For experienced practitioners, the risk is a loss of sharpness, because over-reliance and over-delegation can lead to complacency and declining vigilance. Passive reliance is never a good thing.

(And for organizations as a whole, the risk is that we normalize all of the above, with long term detrimental impacts.)

This matters especially in Procurement because our ability to apply good judgement and make strong decisions that serve the corporate good, even as we grapple with a multitude of competing agendas and demands, is what defines our success. Our work isn’t just about producing outputs but generating better outcomes.

In that regard, AI can help us draft, summarize, analyze, and recommend but it cannot, by itself, deliver real procurement judgment. We still need discernment. We need the ability to read a stakeholder or interpret the nature and magnitude of a particular risk. We need to be able to make real-time, thoughtful, conscientious trade-offs. For example:

• Summarizing a contract is not the same as understanding its comparative risks in context of organizational realities

• Generating a sourcing recommendation is not the same as exercising true commercial discernment that incorporates the nuances of the situation

• Producing a negotiation script is not the same as reading leverage, timing, and context

The fact is that Procurement decisions are often made under ambiguity, across competing stakeholder incentives, with incomplete information and real commercial consequences. So our judgement matters.

And even putting aside the idea of ‘low quality outputs that play at being correct’ (due to hallucinations (still a concern) and errors that will surely diminish as the tech continues to improve), if we don’t own the work, then what is our value?

It’s also worth asking ourselves: If we’re no better than the machines, then why does the organization need us at all?

The Real Question

Look, none of this means cognitive offloading is inherently bad. In many contexts, it’s rational and valuable. The problem begins when we offload the very parts of the work that are building, testing, or preserving judgment.

The real question, then, is not whether we should use AI; that cat is out of the bag. It’s whether we will use it in ways that preserve the foundations and habits that judgment depends on.

Because the danger is not that these tools do our work for us, but that, used carelessly, they can leave us producing the appearance of strong work without building or sustaining the capability to truly own that work.

The challenge for the practitioner, therefore, is to learn how to create music without surrendering musicianship.

The Practical Challenge

That leaves us with a practical challenge: how do we use AI to capture the gains in speed and quality without outsourcing the very cognitive reps that build judgment in the first place?

The answer to that question is multi-faceted - from how we lead on this issue, to how we manage it, to how we train for it.

In a follow-up piece, I’ll outline a practical framework for using AI in ways that preserve cognitive agency rather than erode it

.

In Post 1, I argued that “AI-ready” isn’t a technical identity, more of a behavioral one. This post is the companion piece: how do we put these ideas into practice.

The point of this post is to be practical - not to present shiny tools or some grand transformation program. We have enough of those elsewhere.

My goal is to present a practical way to begin the move from AI-curious (sporadic experimentation) to AI-confident (repeatable, outcome-driven use), while keeping the two things that Procurement can’t outsource: judgment and accountability.

The discussion will be divided into four parts:

1. Common-sense guardrails

2. Practical playbook

3. Useful workflows

4. Getting started this week

First: Common-Sense Boundaries (AKA Don’t Do Something Dumb at Speed)

Let’s start with a reality check.

AI can accelerate your work but it can also accelerate your mistakes. It’s not tuned to do the ‘right things’ all the time, so the onus is on you.

So before we get going, let’s lay down some common-sense boundaries for ourselves. (I know, I know, this is like one of those disclaimers at the beginning of every self help book: consult your doctor/financial advisor/legal professional/etc.)

Here are the guardrails I’d apply at this stage, whether you’re a category manager, an analyst, or a CPO:

1) Treat public tools like glass conference rooms

If you wouldn’t say it on speakerphone in a crowded airport, don’t paste it into a public AI tool. Don’t put any of the following into the public AI tools you use:

• No regulated data (PII, export-controlled, etc.)

• No confidential supplier data

• No non-public pricing, rate cards, rebate terms

• No contract language covered by NDAs

• No customer-sensitive info

• No proprietary strategies or negotiation positions

I know plenty of AI tools have options to protect your data but, for now, I would still err on the side of caution. If in doubt, treat the data as confidential and don’t input it into the tool.

2) Follow Company Policy

Again, this goes without saying, but follow your company policy.

And if your company doesn’t have one, or has a very loose set of guidelines, then assume the strictest stance until your company actually does develop one.

A lack of (or even a loose) AI policy is not permission to do whatever you want, certainly not with company information.

Until you have clear rules, behave like you’re operating in a regulated environment:

• Stick to approved tools only or, where this is no guidance, choose your tools carefully

• Redact all inputs thoroughly and appropriately

• Experiments only with “no risk” and “low stakes” work

• Document everything you do

3) Don’t Treat AI Outputs as “Answers”

AI can be a very fast, very capable intern that displays very high confidence, even as it displays uneven judgment. It will reinforce what you want to hear and will sometimes even tell you things that just aren’t true or valid or right.

So, don’t take it for granted. Don’t outsource your thinking and judgement:

• Verify facts

• Sanity-check logic

• Ask for sources, assumptions, and alternatives

• Pressure-test the output the way you would a supplier claim

Converse with the tool, push back, question its ‘thinking’. At the end of the day, it’s your output and you will be on the hook for it.

4) Keep the Human in the Loop

This is related to point 3, but keep yourself in the loop, especially where it matters - on decision making, judgement issues, etc. If the output affects money, risk, reputation, or legal exposure, then the bar should be even higher.

AI can help you think, draft, and explore but you must still own:

• Decisions

• Communication

• Accountability

Ask yourself: am I comfortable defending this output in front of my boss?

5) Build Muscle Safely

If you’re new to this, don’t start with the crown jewels. Start with non-sensitive use cases that show you the power and capability of the tools.

Start with:

• Meeting prep

• Stakeholder emails

• First-pass research frameworks

• Neutral summaries

• Checklists and question banks

All of these should be more than enough to build confidence without creating risk.

How to Move Up the Curve (And Still Have a Life)

The point of this whole exercise is to get beyond ‘dabbling’ (AI-curious) to ‘standardize’ and ‘incorporate’ into your workflows (AI-confident).

The simplest way I can think of to get there is to make progress without getting overwhelmed:

Step 1: Pick One “Lane” for 30 Days

Choose one part of your job where you want leverage. For example:

• Contracting support

• Supplier intelligence

• Supplier risk insights

• Stakeholder management

Start small. Get results. Embed into your daily work. Expand later.

Step 2: Run Two Reps Per Week

Each rep can be no more than 15–30 minutes:

• Try a prompt (not a one liner, imagine a conversation)

• Produce an output you can actually use

• Improve the prompt next time (”What could I have said/asked that would have given the tool more context/information to have been able to provision a better output?”)

That’s it. I know there are plenty of folks who will tell you to do more and immerse yourself even more deeply - and you can do that. But at least start here. Small reps compound.

Step 3: Keep an “AI Wins Log”

As the saying goes, “If you don’t track it, it never becomes a practice”.

Make a point of tracking what you’ve worked on, what the issues were, what value you saw, etc. You can do this as thoroughly as you like, for example:

• Date / workflow lane

• What I was trying to do

• What I fed the tool (redacted)

• Output I got

• What I changed / validated

• Time saved (or quality improved)

• What I’ll reuse next time

OR just keep it simple: keep a note of what you did, what you learned, what value you recieved and how you could have done better. Make this a personal operating system of sorts.

The point is to capture insights and learn; from “I tried AI once” to “I work differently now”.

Step 4: Define “Better” in Procurement Terms

Stay focused on the practical, tangible, applicable value. Not just “this is really cool output”, but what it means for your work and how you could (and why you should) deploy this on an ongoing basis.

In other words, “Better” means:

• Faster cycle time

• Clearer stakeholder alignment

• Sharper negotiation options

• Fewer risk blind spots

• Better supplier conversations

Anchor your practice to the stuff that you (Procurement) cares about. The more it ‘enables’ you, the better you will be.

Four Procurement workflows where AI can create real leverage

OK - let’s get started.

What follows are practical tasks and patterns you can use immediately, without pretending that AI is some magical, mythical tool.

For each workflow, I’ll suggest:

• What AI is good for

• What you must verify

• A prompt you can reuse

NOTE: I have drafted the prompts to provide guidance for junior as well as senior folks. It goes without saying that if you already have some experience, then tailor this as appropriate to your experience level.

In addition, if any of detail in the prompts below run afoul of the common sense boundaries laid out above, then adjust/edit those, as appropriate.

1) Contracting:

The focus here is on faster comprehension, better questions and cleaner negotiation preparation. The core value is that AI can accelerate your first pass. What it can’t do is replace your counsel or your own scrutiny.

Where AI helps

• Summarize long clauses quickly

• Create a “risk heatmap” of key provisions

• Draft redline questions and negotiation talking points

• Generate fallback language options (as ideas, not legal advice)

What you must verify

• Legal interpretations

• Company- and Jurisdiction-specific implications

• Defined terms and cross-references

• Anything that affects liability, indemnity, termination, IP, data, compliance

Again, AI can accelerate your first pass. It cannot and should not replace counsel or your own scrutiny.

Reusable prompt:

You are a procurement contracts analyst.

My company is a [mid-size buyer] with [moderate] leverage. This is a 3-year agreement valued at approximately $X. We have [one/multiple] alternative suppliers.”

*I’m reviewing a contract for [category/service type] with a supplier. Here are the [redacted] clauses for your review. *****

1. Summarize each clause in plain English.

2. Identify the top risks for the buyer.

3. For each risk, propose questions to ask the supplier

4. Help me identify acceptable fallback positions for risks identified in 3 above.

5. Flag any ambiguous language and suggest how to clarify it.

6. Identify any standard clauses that are missing and explain why they matter

7. Note where any terms deviate significantly from market standard for this category. (If you don’t have market data, label as hypothesis)

8. Flag any clauses that should be reviewed by legal counsel rather than handled by procurement alone

Provide the output in plain-English summary.

Key Note: The goal, as I’ve said above, isn’t “AI reviewed the contract” but that you are able to walk into a legal/stakeholder review with a deeper comprehension and sharper questions.

2) Supplier Intelligence:

The goal here is to use AI to better prepare you for supplier conversations and moving your sourcing strategy forward.

Where AI helps

• Structure and develop a supplier profile quickly

• Turn scattered information into a coherent narrative

• Draft supplier interview questions

• Generate hypotheses about strengths/weaknesses and differentiators

• Build an initial supplier landscape by segment

What you must verify

• Factual claims (revenue, ownership, capabilities, certifications)

• Marketing fluff vs actual valid insights

• Anything that becomes part of a sourcing decision record

Reusable prompt:

You are supporting a sourcing initiative in [category].

Create a supplier intelligence brief for [Name of Supplier (ideally)] or [Supplier Type (less ideal but still workable)]. Do not assume facts. Provide citations (and if you can’t, say so). Label all assumptions.

Include:

• What the supplier likely does well (hypotheses)

• How differentiated these strengths are relative to its competition

• The typical cost structure and where pricing leverage exists for the buyer.

• Common risks in this supplier type

• What creates dependency or switching costs with this supplier type, and how can we structure the engagement to minimize lock-in?

• 12 due diligence questions (commercial + operational + ESG + cyber/data)

• What would make us not choose them

• What should we look for and ask about when requesting customer references?

• What to listen for in discovery calls:

  • Green flags (signals of a good partner)

  • Yellow flags (things that need follow-up)

  • Red flags (signals to walk away)

Keep it concise, bullet-based, and designed for a stakeholder readout. Include a one-paragraph executive summary at the top with a preliminary recommendation or stance.

Key Note: The point here is to use AI to generate structured thinking that you can then validate with real data and supplier calls.

3) Supplier Risk Insights:

AI tools can be great for helping identify early warning signals and develop sharper mitigation plans. The key, as always, is to use them thoughtfully and with your own judgement as central to the analysis.

Where AI helps

• Create a risk taxonomy for your category

• Develop “what could go wrong” scenarios

• Draft monitoring questions and risk dashboard elements

• Generate mitigation options you might not have considered

What you must verify

• Company-specific qualifiers/disqualifiers

• Real-world risk signals

• Financial exposure

• Operational dependencies

• Any recommendation that affects supply continuity

Reusable prompt:

You are a procurement risk advisor.

For [category] with suppliers in [region(s)], create a risk assessment framework.

1. List major risk types (financial, operational, geopolitical, compliance, cyber, ESG, logistics).

2. Rank risk types by severity and likelihood for this specific category-region combination, and explain your reasoning. Not all risk types are equally relevant — deprioritize where appropriate.

3. For each risk type, define leading indicators we can monitor. Suggest specific free or low-cost data sources a procurement team could use to monitor each indicator

4. For each leading indicator, recommend a monitoring frequency (daily/weekly/monthly/quarterly).

5. Create a simple scoring model (1–5) with definitions for each score. For each score level, provide a concrete example relevant to this category so the user can calibrate their assessments.

6. For each risk type, define a threshold score that should trigger an escalation or action, and describe what that action looks like.

7. Provide mitigation strategies (dual source, inventory buffers, contractual protections, audit cadence, etc.). Note those strategies that are proportionate for a contract of [approximate value], and flag where the cost of mitigation may exceed the expected cost of the risk event.

In addition to any descriptive output for the points above, also provide a summary dashboard table (risk type, severity ranking, top indicator, primary mitigation).

Key Point: The point here is not to be exhaustive but to help you identify the breadth of the major risks. You will still need to provide judgment about what’s plausible, material, and actionable.

4) Stakeholder Management:

This is not a flashy use case, but it provides real value in the form of stronger communication, clearer alignment, fewer rework loops, and much faster (and more credible) decisions.

Where AI helps

• Draft crisp stakeholder updates

• Tailor messages to different stakeholder types

• Prepare for tough conversations

• Turn messy meetings into clean decision memos

• Generate options and trade-offs summaries

What you must verify

• Tone and political nuance

• Commitments, timelines, and approvals

• Anything that could be interpreted as binding

Reusable prompt:

You are helping me manage a stakeholder in [function].

Context: [short description].

Goal: [what I need from them].

Constraints: [timeline/budget/risk].

Considerations: [Any history with the stakeholder; key ideas and preferences]

Stakeholder’s influence level: [decision-maker/influencer/gatekeeper/end-user]

Stakeholder’s likely priority: [cost/speed/quality/risk/control].

Suggested tone of communication: [assertive/collaborative/deferential/urgent/relationship-building]

Draft:

1. A 6-sentence email that is clear, calm, and action-oriented.

2. A one-paragraph “decision memo” summary with options and recommended next step.

3. 5 objections they might raise and how I should respond.

4. For each objection, provide the underlying concern driving it, your recommended response, and any phrases to avoid.

5. If I need to compromise, identify the one thing I should protect and the one thing I can concede.

6. Recommend whether this conversation is better handled via email, a brief call, or an in-person meeting, and explain why.

Key Point: The point here is that you’re using AI to remove friction and enhance credibility, so you can spend your energy on judgment and relationship development.

The Difference Between “Using AI” and “Becoming AI-Confident”

At this point, you might notice an underlying theme: none of this requires you to become technical. What it does require is:

• Comfort in experimentation

• Thoughtfulness (and appropriateness) in the prompt structure

• Discipline to verify

• A bias toward turning experiments into habits

• The humility to treat outputs as drafts, not truth

A Short Note for Leaders

If you lead a team, your most impactful move is to make “responsible practice” the norm. You can do this by taking four simple actions:

1. Publish guardrails people can actually follow - what’s off-limits, what requires review, what’s fair game

2. Create a safe space for experimentation - off-limits categories (if any), anonymized data, etc.

3. Reward small, verified wins tied to outcomes - e.g. a better supplier question, a faster risk assessment, a key nugget of insight that moved a conversation or deal forward, etc. - and not “I used the AI tool”

4. Make sharing the norm - via regular forums where people show what worked, what didn’t, and what they learned

The point here is to give your people clarity, safety, and permission to practice.

What To Do This Week

If you’re reading this and thinking, “OK - where do I start?”, here’s one suggestion:

1. Pick one lane

2. Run two reps this week

3. Start your AI Wins Log

4. Share one safe win with someone on your team

That’s literally it - just take simple steps to start becoming the kind of practitioner who can work with these tools, and incorporate them into your workflow.

In a post-AI world, confidence isn’t a result of the tech, but rather your ability to work with it - safely, consistently and with (your) judgment.

Procurement practitioners today are living in what seems like a strange split-screen reality.

On one side, every CPO agenda has “GenAI” stamped on it in bold ink:

• EY’s 2025 Global CPO survey found that 80% of CPOs plan to deploy GenAI in some capacity over the next three years (with more than a third having deployed GenAI in a meaningful way).

• Deloitte’s 2025 Global CPO Survey found that ‘Digital Masters’ (the top quartile performers) are allocating a quarter of their budgets to procurement technology (nearly double 2023 levels) and achieving a 3.2x ROI on their GenAI investments.

The promise (both real and hyped) is there for all to see (and desire) and the exhortations to chase after this ‘promised land’ are ubiquitous.

On the other side, the practitioner’s day job continues to have real teeth: contracts still need negotiating, stakeholders still want their answers yesterday, suppliers still need to be managed actively, while risks continue to show up uninvited. All of these need to be handled now. They require us to leverage all that we know to keep doing “good work”.

But hovering over both these realities is the idea that the baseline for what constitutes “good work” is shifting. The Hackett Group has put an even sharper edge on this:

64% of procurement leaders expect AI and GenAI to transform their roles within five years.

(Note that that’s research from early 2025, before the last year of model releases and product launches that have accelerated the conversation even further.)

And while we can get caught up in discussions about timing, it’s clear to me that the ground is moving. Change is coming.

So the real question isn’t whether AI will matter but, rather, what posture are you taking while AI is becoming the new ‘normal’?

In other words, how are you becoming “AI-ready”?

In this post, I’ll do three things:

1. Explain why ‘learning to code’ is the wrong bar for most practitioners

2. Provide you with a quick self-assessment, and

3. Describe the mindset shifts that move you up the curve.

Why I Stopped Believing “Learn to Code” Was The Answer

For a while, I believed that to be “AI-ready” you needed at least a working understanding of coding. I don’t believe that anymore.

It’s not that I think technical depth is worthless - it absolutely isn’t. It’s just that, for most procurement practitioners, “becoming technically proficient” becomes a treadmill with no finish line. The technology is constantly outpacing us - the models improve, the tools change, the interfaces shift, the jargon mutates, and so on and so on. You can spend months chasing proficiency and still feel behind.

In the meantime, you still have a day job to focus on - a day job that does not and will not pause for the transformation.

Add to this the fact that the “Tech” is also getting easier to use (”the best UI is a text box that you talk or type into”), and it all begs the question:

How do you prep for a new world in a way that advances your understanding and does so in a way that furthers your ability to do your real work more efficiently and effectively?

Or to put it more simply:

How do you become AI-Confident in a way that is meaningful and contextually relevant?

The answer, to me, isn’t technical but behavioral:

• the willingness to experiment (without expecting perfection)

• the ability to ask better questions

• the discipline to verify outputs and not take them for granted, and

• the ability to turn experimentation into repeatable ways of working

In other words: judgment, intention, practice.

The Three States I Keep Seeing in Procurement

This is why I keep coming back to a simple framing. I tend to see three broad types of practitioners: AI-ignorant, AI-curious, and AI-confident.

Let’s dig into each of these types more practically, via a quick self-assessment.

Self-Assessment: Where Are You Right Now?

1) AI-Ignorant (AKA “I’m Too Busy”)

You might be here if:

• You rarely use AI tools unless absolutely pushed to do so

• You think “AI confidence” equals coding or being a technical expert

• You assume the value will become obvious later…”I’ll figure it out then”

• You’re waiting for training, governance, or a corporate rollout to make it “official”

• You avoid experimenting because you don’t want to look foolish or wrong (”I’ve heard it hallucinates”)

• You’re just plain worried

The thing is, a lot of what we call “AI ignorance” is really just passive avoidance - and it’s not sustainable, because the world isn’t going to wait for you to ‘feel’ ready. Sure, the tools aren’t perfect but they are delivering value in different ways today, and they’re going to keep on improving.

So over time, avoidance doesn’t just keep you unchanged, it erodes your credibility because expectations around speed, clarity, and insight will keep rising even if your mindset and approach and workflow don’t. And that’s going to be far worse.

2) AI-Curious (AKA The Default Setting)

You might be here if:

• You follow the headlines and opinions (especially the loud ones)

• You’ve tried the big-name tools a few times

• You experiment for a day or two, then go quiet for weeks

• You don’t yet have a clear sense of where AI helps your work most

• Your exploration is driven by novelty, not outcomes

This is AI-curious, where most people are - and it’s a reasonable place to be, all things considered.

But there is a hidden problem: curiosity without structure doesn’t compound. It remains as a king of ‘sporadic entertainment’, something you “check out” rather than something that you use to change you.

3) AI-Confident (AKA It Isn’t What You Think)

You might be here if:

• You use it with intent, not just when you remember

• You translate experiments into repeatable workflows

• You can point to a few parts of your work where AI reliably helps

• You keep judgment in the loop: you sanity-check, triangulate, and challenge outputs

• You focus on outcomes: faster insight, better options, clearer stakeholder communication, sharper risk visibility

AI-confidence doesn’t mean you can build models, it means you’re focused on constant learning. It means you know where you’d try AI, how you’d evaluate what comes back, and how you would apply it responsibly.

Moving Between The Stages: Mindsets

The really interesting thing about these types or stages is that the distance between them is not as large as it might seem. It’s not a leap from “non-technical” to “technical”; it is, instead a shift from avoidance to dabbling to disciplined use.

So what does moving between these stages actually mean in terms of behavior? How should we think when we make this shift?

From AI-Ignorant to AI-Curious

As I said, this shift - the first and most important one - isn’t about any sort of special intelligence.

It’s about dropping the “wait and see” posture and deciding that “clarity” isn’t something you can wait for. It’s something you can begin to create through low-risk experimentation.

This first step is, therefore, mostly psychological:

• “I don’t need permission to learn.”

• “I don’t need perfection to start.”

• “I can run safe experiments without creating risk.”

It is, therefore, about making the decision to actively learn, to understand what’s out there and begin to understand the potential use cases and value they can deliver.

It’s about just getting started.

From AI-Curious to AI-Confident

Here, you’re adding intent, focus and repetition.

AI-curious folks test tools while AI-confident practitioners build habits - and those habits have a specific flavor:

• They’re anchored to real work

• They’re measured (even informally)

• They improve over time

• They include verification, not blind acceptance

That’s the whole shift - from observation to intentional usage and from dabbling to conscious practice. The key is to move from “AI as novelty” to “AI as a work partner you manage”.

That last phrase - AI as a work partner you manage - matters the most, because in Procurement, as in any other function worth its salt - accountability doesn’t get outsourced to the tool. It has to stay with you.

The Ethos: Take Charge Yourself

If you take only one idea from this post, take this:

Don’t wait for someone to serve AI readiness to you on a corporate tray.

Yes, one day, the training will come, the policies will evolve and the tools will get embedded. There’s no doubt about that. The tech is moving to the point where there will be no option but to.

What you don’t want is for that to be the time you begin learning, because by that time, you’ll have a mountain to climb to simply get to first principles. Worse, by that time, you may be irrelevant.

The practitioner who builds an edge is the one who starts now, safely, and deliberately. And don’t do it because you want to become an “AI person”. Don’t do it with an end destination in mind - there is no end here, it’s a journey.

Do it because you want to stay “high-leverage” as the very definition of leverage changes. Do it because you want to stay relevant.

A Quick Sidebar for Leaders

If you lead a team, your job isn’t (just) to announce that “we are adopting AI”. Your job is to remove the fear and the friction so your people (who are watching you for guidance) can build competence without feeling exposed.

That means:

• create psychological safety for experimentation

• clarify the guardrails (even simple ones)

• reward good judgment, not shiny outputs

• normalize the idea that AI outputs need verification and context

• celebrate small wins that improve cycle time and stakeholder clarity

What’s next

But look, with or without a perfect “tech-forward, people-first” leader, the onus is still on the practitioner. The practitioner still has to choose this posture. It has to be a personal shift before it becomes an institutional one.

So, in the next post, I’m going to suggest ways to get practical. I’ll start with some common-sense boundaries (especially around confidentiality and policy), then lay out a simple playbook to move up the curve, including how to apply AI to procurement workflows where the impact is real:

• Supplier intelligence

• Supplier risk insights

• Stakeholder management

Remember, in a post-AI world, confidence won’t come from the guts of the technology, but from knowing how to work with it - repeatedly, safely and with judgment.

In my last post, I talked about six factors that are driving change across the Procurement function and why, as a result, most practitioners aren’t ready for the future that is unfolding.

For the most part, these factors are the result of developments or pressures from outside of Procurement that are subsequently having an impact inside the function.

In this post, I want to dive into three very specific considerations that Procurement will need to grapple with as it looks to reposition and reinvent itself for a Post-AI world:

1. New Psychology

2. Right Incentives

3. Relevant Investments

Rethinking these considerations isn’t a ‘Procurement only’ problem - it will require support from individuals and groups outside of the function as well. But the impetus for change must begin, as it always should, with us.

Let’s dive into each of these in turn:

1. Towards a New Psychology

When I was with The Smart Cube, there’s a statement I‘d make in discussions with Procurement leaders and practitioners that would always elicit an immediate, visceral reaction - almost always at one extreme or another:

“Procurement’s goal should be to become the biggest value creator in the enterprise.”

As soon as I’d say those words, half the audience would look up, eyebrows raised and energized by the idea, while the other half would, subtly (and sometimes not so subtly), shake their heads, put off by it.

To me, that divergence in reactions encapsulates one of the core issues that impedes the function even today, even after the years of progress the function has made (and it’s got nothing to do with whether you believe my statement - meant as a ‘North Star’ - to be achievable or even realistic):

Do we believe in ourselves and what our mission could be? And do those around us share in this belief?

In many organizations, Procurement is still constrained by this psychology: how the function sees itself as well as how others in the organization view the function.

Of course, history has a part to play here. Procurement has come a long, long way from its ‘back office’ administrator era. We have better tools, more methodologies and a far better caliber of talent across the function.

But that progress is still unevenly distributed, both across and even within companies. Some Procurement teams have advanced, espousing ambitious goals, while others are still in ‘order taker’ mode. At the same time, many Procurement teams are still wrestling with others in the organization (from the C-suite to peers in other departments) who still hold on to the baggage of the “old” Procurement.

Beyond this baggage, there’s also the (rapidly growing) distance between where the Tech is taking the function versus what practitioners have been trained to do. Most Procurement professionals today are built for an era that is rapidly becoming “unfit-for-purpose”, one very much focused on:

• Knowing the process

• Working the data

• Being the ‘expert’

Procuretech - and specifically, AI-First/AI-Enabled Procuretech, is changing that paradigm. As we’ve discussed in prior posts, the model is evolving to one where the prime skills will revolve around:

• Optimal orchestration

• Sound judgement and,

• Delivering better outcomes.

In short, Procurement is moving from being valued for what it knows to being valued for the decisions it enables.

Thing is, we’ve not caught up to this shift, and even where we have in terms of mental acceptance of the idea, we’re certainly not there in terms of wholesale, practical, skills-based readiness.

So to me this is THE foundational issue. We’ve spent decades working to break out of the old psychology and, now, here’s this AI thing, further challenging our identity. That’s as much an emotional and cultural issue as it is a technical one.

2. Towards Better Incentives

If you want the right outcomes, incent the right behaviors.

That’s may be an age-old mantra, but it’s one we all agree on. So much so that we’ve seen significant progress putting it to work in Procurement over the years.

We’ve seen metrics galore: across the function as well as within every sub-function. We’ve seen them in cascading scorecards that encompass both hard and soft metrics. We’ve even seen two-way metrics that cross organization thresholds to include suppliers as well. (If you’re interested, I co-wrote an article for MIT’s Sloan Management review on this topic years ago.)

You would think that with all of this metric sophistication, we’d be in a far better place. Well, we are - and we aren’t. There are a couple of pretty major problems:

First off, most of our metrics today are focused on process and activity. Things like the number of RFQs run, or cycle time or pure savings booked. All useful metrics, for sure, but AI is making these metrics meaningless, because machines will administer a good chunk of the work that many of these metrics inform, to the point that they cease to become “human” metrics.

Instead, we need to focus less on activity efficiency and more on enterprise effectiveness i.e. outcomes (or as close to it as we can get). Outcomes like savings, for sure, but also our ability to:

• Improve time-to-market

• Avoid risk and increase resilience

• Create optionality to manage for unpredictable business outcomes (versus just maximizing realized savings in the moment)

In other words, less of just “we delivered 3% savings” and more of “we enabled a product launch six months earlier by rethinking the supplier ecosystem”.

This mean rethinking our metrics, including expanding our definitions to encompass what constitutes better outcomes and then better measure our ability to help internal customers achieve them.

The second, and to me far more important, issue is that despite the plethora of available metrics, only a limited few are perceived to matter - and, most of the time, just one: hard savings, and more specifically, hard savings that hits my P&L this Fiscal Year.

But this so often leads to conflicting incentives. Not just because of the short-term-ism that it manifests in behavior, but also because better outcomes don’t always manifest themselves in the form of dollars and cents (near-term or otherwise). We’re paying Procurement to be a steward of enterprise value but then we’re not measuring them (not truly) as such.

We need a better and more widely accepted definitions of value - and, yes, that will take more than just Procurement to get sorted. We need the C-suite to buy in as well. That means better ‘customer’ champions, defined quick wins to prove our case, stronger, more coherent communications about our impact, (real) leadership buy-in, and so much more. We cannot go on to ‘self-actualize’ the function unless we do.

Bottom line: Metrics matter. No question about that, but we cannot go on, especially in this Post-AI world that we’re moving into, optimizing for the obsolete.

3. Towards The Right Investments

Ask any leader and they’ll tell you that “people are our greatest asset”. How we operationalize that statement in a Post-AI world, is worth some deeper thought.

Historically, the vast majority of training and development expense has focused on “hard skills” - policy, process, tools, and compliance. And even the soft skills we trained for focused on narrow angles that made sense for a different time:

• Negotiations focused mainly on playing suppliers off against each other and extracting concessions

• Stakeholder management that was more political navigation than proactive multi-party orchestration under uncertainty

• Communications that was more focused on status and updates than narrative construction and management

• Relationship management viewed primarily from a supplier harmony lens than a more holistic portfolio-based perspective

All useful and necessary, no doubt, but sufficient? Optimal?

In an age where the machines will do a good chunk of the work, where so much of our involvement distills down to “dialog with the agent”, and where our engagement moves from engaging with our laptop most of the time to engaging with humans (for tasks beyond fire-fighting), where do we turn our developmental focus?

We need training & development that is focused on two explicit categories:

• Business Judgement Capabilities such as value chain economics, P&L literacy and the ability to make strategic trade-offs. In other words, we need to be able to better assess the business situation at hand and then configure and cultivate the best path to the optimal decision

• Human Leverage Capabilities that go beyond simple concepts such as AI literacy and technical proficiency (both necessary but just part of the overall equation) to also consider relationship management, narrative development, creative problem-solving, the ability to be comfortable and operate with ambiguity, etc.

The underlying point is that we need to cultivate skills that reposition the practitioner as orchestrators and architects. And that’s about more than getting them ready to just execute.

The one other - perhaps even more important - point I’ll make here is that, even with the best tools and intentions, one fundamental issue still remains: when times are tough, training is always the first budget to get cut.

I understand why (in practical terms anyway). Training is flexible and modular, more often than not, seen as a nice-to-have. There’s no immediate system failure when training is paused, no visible outage when human development slows. So it becomes an easy head to serve up when budgets tighten.

But in a Post-AI world, this logic is deeply flawed. Cutting investment in capability building is, in truth, an unstated risk decision, increasing the likelihood of poorer judgment, weaker orchestration, and slower adaptation - exactly when the role of the human matters most.

We need to change this mindset. We keep saying that people are our most important asset and yet we don’t act like it.

If Procurement wants to remain relevant in a world where machines handle execution (and more), it cannot treat the development of judgment, business acumen, and human leverage, as optional.

Cutting investments in capability building in a Post-AI world is nothing less than strategic negligence.

In Summary

The transition to a Post-AI world doesn’t begin with technology. It begins with belief - belief about what Procurement’s mission really is, belief about how it creates value, and belief about whether developing human judgment is a cost or a strategic necessity.

AI will continue to do its thing - compressing execution, automating processes, and surfacing insights, whether we’re ready or not.

What will remain, though, is the distinctly human ability to frame the right problems, make trade-offs under uncertainty, and orchestrate outcomes across complex enterprises.

Procurement’s ability to flourish in this post-AI world means it must choose to see itself differently, align its incentives to that ambition, and, finally, invest accordingly.

Otherwise, instead of leading with value in a Post-AI world, we’ll still be answering questions about the function’s relevance.

In my conversations with CPOs, practitioners and founders over the last couple of years, there’s been a broad consensus around one idea: many of today’s Procurement professionals aren’t prepared for what’s coming.

That is, a sizable portion of today’s Procurement professionals do not have the skills needed to flourish in a Post-AI world. (I fleshed out my perspective on what a “post-AI” world is in my last post.) This is borne out not just in the informal chatter online but also by research:

• Deloitte’s 2025 Global CPO Survey calls this out directly, noting that while a ‘digital‑adept and digital‑capable workforce’ is now a critical differentiator and that next generation skills are needed, more than a third of CPOs cite a ‘talent gap’ as a top barrier to value delivery.

• In a McKinsey discussion on AI in procurement, 43% of CPOs identified strategic thinking as the most critical future competency for category managers, which underscores the clear move (or at least the desire to move) away from transactional skills toward higher‑order capabilities.

How prevalent is this issue? My own (admittedly non-scientific) estimate, yielded from hundreds of discussions over the last couple of years, suggests that the issue is far more serious than even the numbers above might suggest. The figures I hear are as high as 80%, and never below 60%.

In other words, somewhere between 60-80% of practitioners are not equipped with the skills needed to flourish in a Post-AI world.

Why is this the case?

The fact is that most Procurement roles were designed for a pre-AI world. These roles valued compliance, control and standardization as the means to drive efficiency and impact. Organizations and structures were built to reflect this ethos and investments were made in tools and technologies that enabled this agenda.

Then, partly as a result of the function’s success (and you could say its failures), expectations began to change. Spurred on by ever more competitive markets and internal pressures, internal customers and the C-suite began raising their demands of the function. Procurement responded with growing sophistication, led by progressive leaders and a higher caliber talent pool. Combined with the last decade and a half of Procuretech development, we saw the promise, and delivered reality (albeit unevenly) of a different, more value-added Procurement.

AI has now accelerated and magnified all of this.

The net impact is that the expectations of Procurement are shifting faster than the skills, incentives, and operating models can cope. We now are beginning to see a structural mismatch between what the role has trained people to do and what it is being expected to deliver.

Six factors underpin this shift of expectations:

1. Tech

2. Workflows

3. Operating Model

4. CFO

5. C-Suite

6. Internal Customers

Let’s look at each of these in turn.

1. The Tech is Absorbing More Than We Expected

Over the years, with the rise of Procuretech (and most tech in general), most of us assumed these tools would “take the transactional work” but not the “interesting stuff” i.e. all the decision support and cognitive work that humans do.

But, in reality, AI is moving up the value chain much faster than expected.

With the advent of today’s crop of tools (powered by AI), we’re seeing the tech take away work we’ve come to expect as our exclusive domain; not just the transactional stuff but so much more than that. A few examples:

• Supplier discovery becoming an ‘auto-generated’ feature

• Market benchmarks created in minutes

• Cost breakdowns inferred from sparse, public data

• Contract risks flagged before Legal even gets its turn

• Scenario modeling (“what if we dual source?”) done instantly

All of this is decision support work - traditionally a “high value” activity that was part of the developmental trajectory of Junior Analyst to Manager to Category Lead. But with AI absorbing this work, many practitioners are losing not only their learning path but also their value narrative and their differentiation.

Think about it: if you traditionally optimized for producing analysis, you now need to move beyond. Producing the analysis isn’t good enough - machines will do that and you will need to optimize for interpreting, challenging, and owning decisions.

2. Workflows Are Changing Faster Than Skills

If your job is to administer a process, you’re at risk, because AI isn’t just automating tasks, it’s reordering - changing - the way work is done.

Traditionally, we saw workflows that were reasonably sequential:

Intake → Requirements → RFx → Evaluation → Negotiation → Contract

But AI-enabled workflows are different:

Problem framing → Constraint definition → Outcome modeling → Exception handling → Stakeholder alignment

They force us to focus on the bigger picture: What are we trying to achieve? What are alternate paths there? How can I get creative and get to the solution quickly and efficiently and optimally?

In other words, there’s less demand for process administration and a greater demand for judgement, especially on critical projects and at critical moments. The days of “we have a process so let us follow it” are numbered.

So the value premium will be knowing which outcomes matter, where human intervention is necessary, and then being able to figure out when not to follow the process and get to the right outcome.

In the past, we rewarded for procedural rigor, not discretionary judgment. That’s going to change.

If your job is simply administration, coordination, and communication within a process, your job will go.

3. The Operating Model Is Flattening

AI is compressing layers and removing hierarchies.

We’re going to see fewer “Analyst to Manager to Director” handoffs and more direct ownership of outcomes by individuals - because a sizable chunk of the work done in “administering the process” and “analyzing and synthesizing the data” is going the way of the machine.

This means flatter models where visibility increases and output matters more than activity. There will be fewer opportunities (and far lesser tolerance) for excuses. Procurement professionals will be asked:

• “Why did you choose this path?”

• “What trade-offs did you identify and accept?”

• “Which risks did you consciously take or avoid?”

These are questions that go deeper than they might at first sound and the more critical the category (the ones where we want our cliched ‘seat at the table’, the more pronounced they’ll be.

In the past, it was easier to hide behind structure, hierarchy and process, which shielded individuals from accountability and deeper explanation. AI is going to remove this.

4. CFO Pressure will Increase

Given that most CPO’s tend to report to the CFO (and even where they don’t, are very directly influenced by them), the priorities of the CFO are paramount to the existence, focus and future of the function.

If your CFO (truly) believes in Procurement, the world is your oyster. But if he or she doesn’t, then Procurement’s role will be relegated, for the most part, to the transactional. The advent of AI tools has only exacerbated this.

Because CFOs don’t see AI as just a “nice to have”, they see it as a capacity unlock - though perhaps not entirely in the way that the progressive practitioner might want or expect.

CFOs are, rightly, asking questions such as “If intake is automated, why do we need as many buyers?” or “If sourcing cycles are faster, why isn’t throughput higher?” or just simply “Why do we need as many people in Procurement?”

And to be fair, the answers given the current pace of technology are “We don’t”, “It should”, and “we don’t”.

AI is creating an expectation, initially implicit but becoming much more explicit: Headcount should not scale linearly with workload anymore. In fact, we are likely past the point of peak employment in the function.

In other words, if we have any hope of staying relevant, AI demands that practitioners clearly articulate business impact every step of the way.

So we have two options: Either we accept this and the functions turns into a distributed, self-serve, transactional driver of spend, or we upskill and focus on becoming value creators.

If the hope is to stay relevant, there’s only one answer.

5. C-Suite Expectations Have Shifted

In 2024, the main question from the C-suite was “Are we using AI?”. Moving into 2025 and now 2026, the question has shifting to asking “Where is AI changing outcomes?”

In Procurement’s case, this translates into tangible outcomes:

• Cost reduction (of course)

• Faster time-to-market

• Measurable working capital impacts

• Supply risk anticipation and timely mitigation

• Margin protection under inflation

• Innovation through suppliers

The point is that we’re moving beyond AI theater, beyond the simple reporting to the Street about all the cool initiatives we’ve implemented, to ‘here’s what the new tech has actually delivered and changed’. Procurement professionals, for their part, are increasingly being expected to:

• Identify where AI should be applied

• Define clear success metrics

• Explain why its use case matters commercially

This means a shift from executing initiatives to shaping them - a skill that will be the domain of not just a few senior folks but that of the entire (smaller but more capable) function.

6. Internal Customers No Longer Tolerate “Generic Procurement”

Some practitioners bristle at the idea that they have “customers”. They say, instead, that “we have partners, we are equals”. This is an argument around semantics and misses the point. Procurement exists to serve the business, which has asked it to manage a specific set of spend to better meet the goals of the overall business.

In other words, Procurement has customers. And those customers are demanding a much higher caliber of performance - especially as so much of a corporation’s work is “AI-ified”. Their key question is:

“If you cannot deliver clear and specific value to my BU or department, then I don’t need you.”

Think about it. If AI can run an RFQ using simple chat prompts, conduct the analytics, benchmark pricing and even flag Contract-to-RFQ risk, then it’s right for the business to question the value of a Procurement model that cannot meet it’s specific requirements.

This means a shift from historical emphases on ‘standardization and scale’ to ‘contextual, custom relevance’.

The real value drivers, then, are category intimacy and business understanding as well as the ability proactively spot opportunities to create custom value.

Ask yourself, how many of your team are ready for this future?

In Summary

The above, in my view, are the prime reasons why Procurement is changing and why the function, as a whole, is not quite ready for the shift.

(In my next post, I’ll speak to some of the internal structural and historical factors within Procurement worth considering as we work to engineer the shift for ourselves.)

The net of all this is that, if we’re not prepped and ready, then all of the concerns we have about “the machines coming to take our jobs” and “Procurement becoming irrelevant and being phased out of existence” will come to fruition.

And this won’t be because of a lack of intelligence or work ethic on the part of the individual practitioner. It’s got nothing to do with that. It will be because AI is eliminating the version of the function that we have historically optimized for.

And, in my mind, if we let that happen, then it will be a problem not only for the function and its Practitioners, but for the corporation as a whole.

We will lose the human element. We will lose the ability to truly manage through ambiguity. We will lose accountability.

In the end, we will lose as a corporation.

The “Post-AI” world.

I’ve used this term a lot and often get asked what I really mean - not just what it is, but when it will arrive, and what the world will look like when it does.

In this post, I’ll parse through exactly what I mean - and also why we need to move beyond the conventional interpretations of the term.

AI is Already at Work

There’s plenty of debate about what AI is, it’s “true value” and what it means for the work we do. There are even conversations about whether AI is “real” and if, in time, it’ll end up being just another simple tool, like so many other tech solutions out there.

I’ll start by saying that the bus has left the station. AI is real and it’s already delivering value.

Individuals are using it at work to do their research, think through strategies, and approach new business interactions in thoughtful ways. (And that’s not even touching personal use cases outside of the workplace.)

Companies are using AI as well to automate transactional processes, implement agents to take on specific tasks (reasonably) autonomously, and accelerate time (and path) to outcome. (Obviously, there’s much more work to be done here, but the trend is positive.)

So, sure, we can debate the nuances of AI all we like, but the fact is that its value is here, at the very least, in its early stages.

AGI is a Red Herring

Another discussion that you hear a lot in conventional media as well as the social media sphere is about Artificial General Intelligence or AGI - an all knowing technology that will do everything we humans can do but better.

Personally, I think all this talk of AGI is a red herring. No one (really) knows if it’s even possible or, if it is, when it might become a reality. So, for all practical purposes, until we know substantially more, it doesn’t matter for what we do.

(Personally, I don’t believe we’ll get to a perfect replication of the brain in machine form, at least without severe deficiencies and/or material adverse consequences. Our brains are just that complex. You can fight me on this later.)

In other words, AGI as ‘full autonomy’, ‘zero humans’, etc., is just a talk track. Let’s not waste cycles on remote eventualities. It just delays preparation for what already matters.

Instead, let’s focus on tech we can actually work with. The AI of the here and now.

Post-AI is Not a Future State

When I say “Post-AI”, I don’t mean when “AI is finished and perfect”.

I mean when “AI stops being optional”. It ceases to become a differentiator, it becomes an assumption. It’s when AI is embedded, imperfect, unevenly deployed, constantly improving and already reshaping expectations. It is NOT a future state waiting for the perfect tech, sometime in the future.

It is, in actual fact, now. It’s our current reality in which AI is already embedded, one way or another. Imperfect, yes, but still immensely valuable.

In this reality, we’re already seeing the signs in Procurement - a trendline moving away from a process and task focus to one where Procurement roles are being reshaped around outcomes, orchestration, and human judgment.

In this world, advantage accrues from better roles, better decisions, and better interactions - not (just) from adopting more AI.

The Four Pillars of Post-AI Procurement

So my point is that the post-AI world is now.

And to make sense of it requires us to understand it’s multiple facets. Not just “AI as Tech” but rather, its four foundational pillars: Technology, Focus, Operating Model and Mindset/Human Capability. These pillars are the lenses that help us frame this world we’re now dealing with.

Let’s look at each one in turn.

Pillar 1: Technology

Yes, the tech is an essential ingredient. But in a post-AI Procurement world, the tech doesn’t need to be perfect, nor does it need to be everywhere. It just needs to be intentionally and precisely deployed.

Don’t worry about how advanced it is, just focus on using it thoughtfully:

• Consider narrow sourcing use cases that accelerate outcomes

• Recognize its value as a tool for decision support (vs decision replacement)

• Treat the need for the “Human-in-the-loop” as a feature, not a bug

Pillar 2: Focus

Shift Procurement’s focus from its pre-AI emphasis (compliance, consistency and process fidelity) to business outcomes and internal customer satisfaction. This is, of course, harder than it sounds, because it’s weighted down by myriad factors, which we’ll get into in future posts.

The point is to focus on delivering the right results for the business. Stakeholders don’t care how Procurement gets there, they just want to know:

• Did you help me hit my goals?

• Did you reduce friction i.e. did you make the process as easy as it could be?

• Did you improve the final decision aka ‘did having you in the room matter’?

Pillar 3: Operating Model

We need to rethink our operating models (how we deliver) from “empires” to “orchestration”.

AI breaks the logic of large centralized teams, hierarchies and the hoarding of capabilities. Instead it demands that we move towards a modular mix of small cores of accountabilities, optimal expertise (both internal and external), carefully considered Make vs Buy decisions, and the liberal (but thoughtful) use of Agents.

Practically, this means:

• More rapid movement from requirement to insight to decision

• Less ownership and, maybe more accurately, administration of the tools

• More ownership of outcomes (which requires much clearer accountabilities) and,

• Fewer fiefdoms (because Procurement’s scale comes from orchestration not from headcount).

Pillar 4: Mindset and Human Capability

In the post-AI world, outcomes are all that matter.

Which means judgment, trust, and business fluency will matter more than ‘working through the system’. Which means technical differentiation will matter far less than human differentiation.

Post-AI Procurement professionals will value judgment under uncertainty, stakeholder influence, commercial intuition (aka business sense), narrative and storytelling as well as sound reasoning.

Soft skills, more than anything else, will become differentiators as Practitioners thrive by generating options, helping stakeholders choose better paths, manage for the right trade-offs and own the outcomes (and their consequences).

Post AI is Already Here

So, to me, we’re entering the Post-AI world, if not already very much in it.

The tools and technologies to help us are already out there. They may carry different labels - AI-First, AI-Native, AI enabled - and they may be imperfect, but they’re already changing expectations. What matters now is how we thoughtfully adopt them.

And, as I’ve outlined here, this will be about so much more than the tech itself.

It will be about our ability to rethink the very pillars that define the work of the Procurement Practitioner - delivering better judgment, clearer accountability, and a redefinition of what it truly means to add value.

AI will keep improving either way. Whether Procurement adapts its roles, models and mindsets fast enough to stay relevant is the real question.

One theme that keeps resurfacing in conversations about AI in Procurement is that the technology is advancing far faster than the practitioner. And that gap matters far more than most tool-level debates.

I explored this tension and why the human must remain at the center of any credible AI roadmap in a recent conversation with Shaun Syvertsen of ConvergentIS. We went beyond surface-level hype to examine what this shift really means for how Procurement works and how practitioners lead.

We covered a host of topics, including:

• Why today’s practitioner needs to adapt to the Procurement of the future

• What we can learn from software development in terms of AI adoption

• What Procurement leaders need to remember as the model shifts

You can watch the full discussion below:

There’s a quote doing the rounds that I keep coming back to:

“I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes.”

In today’s procurement world, that sentiment feels salient.

AI is here and, all around us, we’re hearing about how it’s going to change everything we do: next-gen our category management; source more and better - mostly with and for us (but sometimes without us!); help us analyze our contracts and identify massive potential; drive far better supplier management; and on and on and on.

But many practitioners are having an ‘existential’ moment.

Instead of seeing AI as freeing us up to do the creative, strategic work we’ve always wanted to do, it’s making us question what we even can do anymore - or worse, whether we’ll even have a job to do in a few years?

(To be fair, a good number of folks in Procurement have been having such conversations for some time - with a few very progressive thinkers even questioning whether (and how) the function can stay relevant in the (not too distant) future.)

I think the answer to this question lies in the fact that AI is not just another tech wave. This isn’t like e-sourcing in the 2000s or cloud migration in the 2010s - all fantastic developments that, to be sure, changed the game for practitioners and providers alike. AI is deeper than that.

The current wave of Procuretech being developed and delivered promises to change the way we think about and do our work, to an extent that I’m not sure many have come to grips with.

And so they react in the ways humans always react in these sorts of situations: they blank it out and keep plugging away, or they get worried and have an existential crisis.

Or they decide to do something about it.

This post (and my new site, Proquria) is my opening salvo to get to the bottom of what we can do about it.

The Buzz (and the B.S.)

Look, I get it. We’re in a hype cycle. I can feel it. You can feel it.

New product releases just about every week. Sky-high valuations. LinkedIn flooded with “AI-first” everything. SaaS platforms claiming to be “AI-enabled”, “AI-native”, “AI-augmented”. VC investments breaking records. One-person companies building agents that do everything under the sun. Talk of AGI either coming for all of us or coming to save all of us (I can’t quite tell which).

It’s all overwhelming.

Conversations have moved from automation to “agentic workflows.” Traditional service providers are suddenly, now, AI platforms. Boutique firms are calling themselves AI consultancies. Everyone’s rebranding.

But, beneath the buzz (and, yes, plenty of BS), something real is happening:

• Autonomous sourcing agents are generating RFQs, inviting approved suppliers, analyzing responses and recommending optimal award scenarios

• AI-powered tools are reviewing SOW’s against RFPs, highlighting negotiation areas, redlining contracts and prepping all outgoing comms

• Supplier Risk monitors are leveraging real-time data to monitor risk factors, scoring suppliers and suggesting mitigation actions

This new breed of Technology (which, by the way, isn’t just about Gen AI or agentic systems, but encompasses rules-based automation, workflow orchestration, traditional RPA, and decision support tools) promises to restructure the way Procurement work is done.

We can’t ignore it. In fact, we have to expect it.

If we look at how Procurement work is structured, in a simple sense, there are three layers:

1. Strategic activity (strategy development, supplier partnerships, stakeholder alignment)

2. Cognitive or Decision support work (running RFPs, implementing strategies, data synthesis and analysis)

3. Transactional work (PO processing, invoice matching, triage, etc.)

The Procuretech revolution of the last few decades has shown us that the transactional work is going the way of the machine and AI is simply accelerating this trend.

But what’s new is that AI is now promising to take on a ton of the work in bucket 2 (and some of bucket 1) as well. Decision-making is being expedited and better informed by Generative AI. AI can simulate/guide negotiation outcomes more fruitfully. It can even provide guidance around alternative category strategies.

ROI Lags (For Now)

That said, we’re still early.

For all the excitement, many organizations are still struggling to get past the pilot phase. Leaders talk about AI strategy, but most deployments are siloed, underfunded, or misaligned with how the function actually operates. Here’s what I’m seeing:

• Top-down mandates: Executives want AI efficiencies, but often don’t think beyond implementing tools

• Vision issues: Teams trying to create or stitch together solutions without clear maps, ownership or training

• Overhyped expectations: Trying to go big too fast leading to underwhelming results

• Security concerns: Data governance and confidentiality fears that delay adoption

• Buy or Build: The perennial debate about DIY versus buying expertise. all he more prominent given the rise of vibe coding

• Lack of skills: Practitioners who are AI-curious, not AI-confident, and then limited by some or all of the above

For all of the above reasons, the short-term payoff isn’t clear yet.

Long-Term: Still Wildly Underhyped

BUT, for all the noise, the long-term potential of AI in Procurement is still underhyped, because the nature of Procurement work makes it incredibly well-suited for intelligent systems. From supplier discovery to risk modeling to contract intelligence, the opportunities are enormous - and the value is massive.

Done right, AI can help the function:

• Shift us from ‘Spend Policers’ to ‘Value Orchestrators’

• Free up time to focus on human-centered challenges such as stakeholder influence and SEI

• Serve as true advisors to the business

But, of course, we won’t get there just like that, and it won’t be because of the tech.

It’s Not About the Tech

The biggest barrier is us.

It isn’t about the tech stack, it’s about the practitioner - or to temper the subtitle to this post just a little:

We need to reshape the practitioner, not just the Tech stack.

As AI becomes more powerful, expectations of Procurement will only increase.

• The C-suite will demand more: more insights, more speed, more “AI-ification”

• The CFO will expect sharper business cases and fewer headcount asks

• Stakeholders will want smarter, more embedded support

Meanwhile, the function itself will (be expected to) shrink - at least in terms of headcount. (I’m not the first person to say this but we are likely past peak employment in traditional Procurement roles.) Tools will become easier to use. Platforms will be abundant. Plug-and-play options will proliferate.

So what’s going to matter will be the human at the center of this operating system. And that human will be a results-focused architect and orchestrator.

Are Practitioners Ready?

In my informal conversations with leaders and practitioners, there’s a shared concern: many of today’s Procurement professionals aren’t prepared for what’s coming. Estimates (non-scientific, admittedly) for how many, range as high as 80%, but never fall below 60%, at least in my conversations.

And this goes beyond simple concepts such as AI literacy and technical proficiency, which is just one part of the equation. It’s also about:

• Relationship management

• Narrative building

• Creative problem-solving

• Comfort with ambiguity

• Enterprise thinking

• New expansive definitions of value

• And a host of other factors that influence our ability to become the orchestrators and architects of the future - including alternative definitions of the Procurement “org chart” and operating model (including the end of “fiefdoms”)

In other words, what is the practitioner going to do when there’s no more admin work and far less fire-fighting? Are you really ready to optimally do all of the above?

So What Now?

This isn’t a call to panic. The good news is that we’re still early. The transformation is still unfolding and, honestly, no one is an expert. We’re all learning at the same time - the pace of innovation ensures that will be the case.

But it is coming and we need to be ready if we want to stay relevant, as individuals and as a function.

That’s what Proquria is here to explore. It’s a call to prepare, and also understand how.

So, let’s start asking better questions. Let’s figure out how we prepare the Procurement practitioner, and not just the tools, for the brave new world ahead.