“The Suite Was Supposed to Settle Everything.”

This is a requested response to Paul Marty’s inquiry: Is Source-to-Pay Dying.

We’ve heard the “one app to rule them all myth” since the late 60s when IBM said the MRP would control the operations and no other app would be needed. This was rekindled in the mid 80s when Oracle said the ERP would control the organization and no other app would be needed. Then in the late 90s when the web would provide a common interface to everything. Then the late noughts / early tens when the Suite would solve everything. It’s the myth that keeps on giving.

The reality is that any new monolith application model only solves all the problems that came before, not the problems you have now, and not the problems that are coming.

That doesn’t mean suites don’t have value, or that suites won’t provide more value than any other alternative, just that it’s not a panacea. Determining whether it provides more or less value requires understanding the true total value realized (taking all costs and benefits into account) compared to the true total value of the alternatives, many of which are unproven.

“Is The Economic Value Moving Somewhere Else?”

First off, let us note that, fundamentally, the new AI orchestration craze is no different, as it ultimately depends on a protocol that can bind everything together. This isn’t new either. Remember CORBA? Probably not, as most of you probably don’t even know what it is. Look it up. At the dawn of the dot com era, it was going to solve everything. It made all the same promises MCP is making now.

The reality is that the fundamental truths of technology have not changed.

> If it sounds too good to be true, it is.

> Technology can’t advance any faster than discovery (either mathematical computing theory or practical scientific discoveries that allow for new engineering that can build better machines to support better models), and that has, and will, always be slow. How many hundreds of thousands (or millions of years) has it taken us to compute a theoretical warp bubble, and how many decades/centuries more before we can achieve it? (We’ve went from conjecture with the introduction of Star Trek 60 years ago to theoretical possibility 5 years ago … but now we need to create the energy source, the materials that could withstand it, etc. etc. etc.)

> … businesses not only need to make money every year, but are punished by the capitalistic economy if they don’t; so even if all they have is the same old sh!t, they still need to sell it, and the best way to do that is to slap a new coat of paint on the old Mustang and tell you it’s a brand new model (even though the engine hasn’t changed in almost a decade)

This doesn’t mean that AI orchestration doesn’t provide value, just that it’s not the ultimate answer and that you have to truly understand it’s true cost vs its true benefits and, where autonomy is granted, true risks before you can make a decision.

“What if the employee does not need to enter the procurement system at all?”

That doesn’t change the fact you still need a procurement system.

Just because an intake layer can work across applications, and just because an AI agent can complete a routine workflow without ever displaying the myriad of screens suite vendors spent years (if not decades) refining, that doesn’t mean that the core functions go away. Either those have to be part of the intake/orchestration system (in which case it’s really a suite with a new kind of interface) or you still need the underlying best-of-breed / suite systems, which are only accessed through their APIs.

In the latter case, the true economic value remains in the best-of-breed/core systems … and the only value of the intake/orchestration/AI agent is the cost of user time reduction (i.e. 5 FT AP agents to manually process all the invoices vs 1 FT AP agent to manually deal with the exceptions). So if the cost of the software is greater than the manpower saved, it’s negative value.

This assumes that the agent is deterministic and will do all processing according to user defined rules and in user defined bounds. If you are using LLM-powered agents and giving them autonomy, there is no guarantee they will follow all rules and stay without bounds — autonomy means that they can decide only 10K over their approval limit is ok because it appears to be a known supplier, onboard an uncertified supplier, issue a new PO to a supplier who is late on other critical deliveries, etc. which can result in significant losses.

Which brings us to the most important question:

“ONE INTERFACE > MANY SYSTEMS > WHO DECIDES?”

This is a key question for multiple reasons:

1) You can’t cost / benefit a system without knowing.

As above, if the decision is ultimately human, you can accurately cost the system and determine it’s fixed (workforce reduction) benefit as well as it’s expected benefits from (weighted) average results and track true benefits over time.

But if the decision is ultimately the system, you can’t truly estimate the benefits as you don’t know what it will do, how many mistakes it will make, and what the cost to clean them up will be — and if it uses LLM technology, you can’t even cost the system as you don’t know how the token/compute cost is going to increase over time as you don’t know the token growth that will be required or how soon you’re going to have to pay true compute costs (the big AI companies can’t keep losing $4 for every $1 of revenue, and with today’s energy and water costs, the true compute cost is about 5X what most of the players are charging … and this will only increase over time.)

2) You can’t quality control without absolutely knowing who decides what, what the range of options are, where, and how to encode the checks.

Which means that:

“THE CENTRAL QUESTION … IS WHO OWNS THE REASONING THAT DETERMINES WHAT PROCUREMENT SHOULD DO”

Especially when the answer to:

“Are Twenty Agents Really Twenty Opinions?”

is NO!

I know Jon W. Hansen might disagree with me here, since he uses multiple systems to arrive at his (starting) algorithmic analysis of the marketplace regularly, but agents reason off of the data they are given.

When you interconnect them, they share “data” and “reasoning” continuously. Also, even though they are ultimately sophisticated probabilistic models no human can fully understand, the core is similar to deep multi-level feedback neural networks. This means that common “data” and “reasoning” increases the weighting on that “data” and “reasoning”, increases similar output, which increases similar input to the minority models, which causes that “data” and “reasoning” to become more heavily weighted in them. Or, in simple terms, a majority of agents will bully the minority of agents into their viewpoint until there is one opinion, not twenty.

Even if all 20 agents start out different, as they share “data” and “reasoning”, there will be some commonality, and as the “data” and “reasoning” flows back and forth, the weighting of the commonality will increase until it snowballs into a majority, given enough exchanges (over time).

While some “experts” will try to tell you this all but certainly guarantees you will get the best results once the agents agree, that is ONLY true if the common “data” and “reasoning” is accurate. Some agents will have bad / fake / hallucinated (wrong) data and others will share faulty “reasoning” because they don’t actually reason. (They predict arguments … think of them as a super powerful fancy version of next word suggestion on your iPhone, except they are predicting answers to questions and sometimes even the steps involved to get there!)

While individual agents sometimes work well if they have enough raw question/answer input/output previous step/next step data top work from (and that’s how they create mathematical “proofs” — they can scan millions of research papers and proofs, build up all the associations as input/output pairs, and then see if there is a way to chain them all together from a question to an answer where the output of one step in one paper is the required input of a step in a paper in another paper, possibly published in an unrelated journal in an unrelated discipline), they sometimes don’t work very well at all!

They’ll tell you to eat one rock a day. Or that strawberry only has 2 rs. Or, because they don’t understand reality, when asked, they will tell you that “the woodchuck could carry 57,600 pounds of wood in a day” when asked “How much wood could a woodchuck chuck if the woodchuck was 10 lbs?” (I didn’t make this up, DeepSeek did!) (And, even worse, it gave some extremely ridiculous “reasoning” to back it up which might fool someone who didn’t think twice because the logic was pretty good, while the data it worked on was ridiculous [and obviously hallucinated]. I should also point out that this was a result “I” totally expected because there shouldn’t be much “data” on “how much wood a wouldchuck could chuck if a woodchuck could chuck wood” on the internet [which is it’s training data], because we all know a woodchuck can’t chuck wood but the AI doesn’t know that.)

Thus, whether you use one agent or twenty, you will get one opinion, and one result, which could be a total hallucination if none of the systems had enough correct “data” and examples to “reason” on.

Given that, as Paul Martyn points out, the best decisions are made when there is

  1. A diversity of experiences and information to draw on
  2. Independent thought by the participants (that can provide truly different views to the ultimate decision owner)
  3. Decentralization of knowledge for risk mitigation

and that today’s AI, which is NOT intelligent, cannot be trusted to always make the best decision as

  1. all agent systems have access to the same data on your network and the web
  2. agents cannot think, only produce “outputs” based on different weightings of “inputs”
  3. agents with minority data/reasoning can, and will, have at least their reasoning, if not their data, overwritten by the majority

This brings us to the following conclusions:

1. Traditional Suite and BoB solutions ARE still needed, even with orchestration and agents (especially since the later should be restricted to either fully deterministic processes or processes where the cost to fix/clean up after an occasional bad decision by software is less than the cost of having a team of people do that process manually)

2. The New Hotness Hype will always take sales away from established players, as the lazy and greedy look for that panacea solution that does not, and cannot exist … but when those solutions fail to achieve the result, some of those sales will return (as the only other option will be to build the missing parts into the acquired system, and why waste money trying to extend a system that’s fundamentally broken?)

3. “The Market is Not Dying”, as Paul Martyn states, but it is shifting. Suites will become the transaction engines and systems of record in organizations that experiment with new interfaces, orchestration, and AI agents. Hybrid suites, which embed orchestration and intake, will take over in organizations that want modern connectivity and interface but want to limit the number of vendors they need to deal with. And traditional suites will gain traction in highly regulated industries and governments where everything has to be 100% predictable and in-line with thousands (if not millions) of pages of policy regulations!

As Jason Busch likes to say, “many things can be true at once” and as long as those things are not mutually exclusive, they will be in the market.

(P.S. Thanks for taking up my day with no warning, Paul!)

Is Global Sourcing Still Worth It?

Think about the baseline requirements.

  • Supplier Selection
    • potential supplier identification
    • verification and risk analysis
    • onboarding
  • Purchase Order Generation
    • order creation
    • transmission
    • acknowledgement
  • Route and Carrier Selection
    • potential route identification
    • potential carrier identification for remote trucking, ocean/air freight, local trucking
    • final carrier selection
    • cross-dock identification
    • FTZ identification
    • route selection based on cost and network balance
    • carrier selection
  • Transport Insurance
    • multi-quote obtainment and analysis
    • insurance provider selection
  • (Inventory) Financing
  • Trade Document Creation
    • export document creation
    • (automated) cargo manifests
    • import document creation, inc. security filings
  • Transport & Shipment Tracking
    • loading
    • transit
    • cross-docking
  • Importation
    • holds
    • inspection
    • tariffs
  • Documentation
    • goods receipt
    • invoice acknowledgement
  • Payment, Taxation & Reconciliation
    • payment authorization
    • pre-clearance
    • tax payment
    • supplier payment
    • refund tracking
    • refund collection
    • reconciliation

These are just baseline. Even the most simple global sourcing situation is insanely complicated. On top of this:

  • supplier risk analysis is insanely complicated as you have to consider all the supplier risks and supplier location risks
  • … as well as the raw material risks that affect their ability to acquire the materials and components they need to produce their product
  • carrier risk is also complicated as rapid fuel increases or primary lane closures that cuts off a significant portion of their revenue can take a financially healthy carrier to an almost bankrupt one
  • tariff risk is now a thing as one major country in particular changes tariffs on the whims of its leader; outdoing the chaos Brazil charged when it was changing tariffs weekly
  • insurance risk is real as the increasing risk of natural and man made disasters has resulted in insurance rates going up
  • every year, even though you’re paying more for insurance, more and more exclusions are being added with conflict, war, and piracy on the rise (again)
  • export, manifest, and import documentation, as well as tariff filings, are becoming insanely more complicated by the year … with every country and carrier having different requirements
  • the countries with e-invoice requirements (clearance and post-audit) tend to invent their own, slightly different, requirements and dealing with global invoices from 100 countries requires 100 different rule sets — that’s why there are vendors that just do global e-invoice compliance
  • global e-payments can be complicated as well … not all countries support your favourite e-transfer platform, accept your primary bank’s payment network, or deal with your preferred currency
  • VAT/Taxes can be just as difficult to pay, and ten times more difficult to reclaim

So, if you can get the stuff locally, or next door (i.e. one country over), why should you go global?

Especially when the cost argument doesn’t cut it anymore. Even if the model still comes out making the global source cheaper, the cost of risk, which needs to be taken into account as the potential of any multinational experiencing at least one major disruption a year is approaching 100%, means that the losses in any category will eventually exceed the savings (because the chances of a disruption for any major lane or supplier increase over time — you can’t treat it as an independent event because even coin flips aren’t independent — one is 50/50; but if 9 in a row have been heads, the chance of tails is almost 98%).

In other words, if it’s not necessary, does it still make sense?

Managing Your Relationship Off of a Scorecard? I give you an F!

Unless it’s a two-way scorecard, then I give you a C- for effort.

Don’t get it wrong, SI likes two-way scorecards. It’s the only scorecard ever developed for Procurement that’s any good. There’s no other way to properly capture subjective opinions which can be just as important as objective relationship measurements. If there are misunderstandings or personnel issues, a focus on improving processes or efficiency based on objective measurements will be all for nought since whichever side believes the problem lies with the other will be adverse to making any effort (and if you’re only capturing one side, who knows if you’re capturing the right side).

But many SPM platforms define the (two-way) scorecard as their entire foundation of relationship management when it’s just a minor tool. A very important tool in the toolkit … like the measuring tape that a carpenter shouldn’t do anything without — as no relationship manager should be without a two-way scorecard, but still a tool.

But that’s just a tool to measure alignment. You need to focus more on ensuring there’s alignment to start with. This means you primarily need to:

  • ensure you select the right supplier in the selection process
  • clearly define what you expect from the engagement
  • define milestones and key metrics you will be monitoring and agree on the definitions
  • regularly meet to discuss improvement, stagnation, decline

ensure you select the right supplier in the selection process

Your selection process must ensure:

  • you vet them for any risks or red flags that go against your policies
  • you clearly define the product / service requirements
  • you analyze their offering against those requirements
  • you analyze their culture

Frankly, you’ll never have good performance if

  • the relationship ends up being short term because you missed a red flag that legal or risk management won’t put up with
  • you don’t define precisely what you need in functionality and service level requirements
  • you don’t ensure the supplier’s product or service isn’t a great fit for you
  • their culture doesn’t mesh with yours

clearly define what you expect from the engagement

It’s not just saying you want a product that does A through ZZZ or a(n outsourced) service that provides you a detailed transformation plan, takes over your procurement function for tail spend, or helps you onboard your suppliers. It’s a

  • detailed description of what functions you are looking to transform, whether you are just looking for process redefinition support or the selection and implementation of new technologies and automation platforms, and how much project management and assurance you need
  • detailed definition of what tail spend is in terms of categories, amount, frequencies; what type of tail spend can be catalog/punch out, which should be quoted, and when it should be passed back and included in a (modified) category or treated as strategic spend; along with detailed platform requirements, the data collection and model requirements (for integration with your master data and analytics systems), and whether or not it’s okay to use a GPO
  • what processes you use now, the data required for every supplier as well as specific data requirements by category or region or spend, who is responsible for support during on-boarding and post-on-boarding and for what, what systems should be integrated with, and so on.

Just like there’s no success without clear requirements, there’s no foundation for a good relationship without clear requirements of expectations.

co-define milestones and key metrics you will be monitoring and agree on the definitions

In order for a party to perform well, they need to know precisely

  • what they are being judged on
  • how the metrics are defined
  • what the ultimate goals are

The reality is that what get’s measured gets managed, and the management that happens is that which aligns the performance to what improves, or at least maintains, the metrics — whether it is the right thing to do, or not. For instance, if your supplier is hooked on your early payment discount offering as a form of financing (it’s not, by the way, but that doesn’t stop most organizations from treating it like it is), and you make that availability contingent on a 95%+OTD metric, and every shipment that arrives the next calendar day, even if the difference is 4:00 pm for end-of-day unload vs 8:00 am for start of day unload, is counted as late, your supplier is going to do everything it can to not only be on time, but attempt to get all deliveries out early to make sure construction, weather, or unexpected delays don’t stop the shipment, even if that means paying a trucker to sit in a parking lot with unloaded inventory for 2 days! And if getting it out that early requires factory workers to go from 10 hour shifts to 12 hour OT shifts, which makes them tired and increases error rates after hour 9, so be it! End of the day, you’ll get your OTD and pay more for delivery and have to deal with higher defects and waste and all of the tracking and return costs incurred. If the goal in introducing OTD was just to minimize customer stock-outs, you didn’t exactly achieve it because if you end up having to reject a (considerable portion of a) shipment due to poor quality, you’re stocked out (and the OTD metric gives them a pat on the back).

This is why we said “co-definition” — both parties should work together to define (blended) metrics (and scorecards) that will not only achieve the goal, but allow each party to be measured fairly.

regularly meet to discuss improvement, stagnation, decline

Emailing your counterpart a scorecard and saying “You did great here, acceptable here, but sucked in these parts — Do Better” is not relationship management because it doesn’t even imply any sort of relationship. Relationship management is developing the collaboration and desire for both parties to work together, not just pointing out deficiencies in a scorecard and saying “do better”.

All of this means that if you want to do true relationship management, you need to build in relationship management into everything you do because just defining a scorecard-driven program after the fact when the supplier performance isn’t too your liking isn’t relationship management.

How Do You Execute Effective Supplier Performance Management in Today’s World?

This is an interesting topic. Many books have been written about this; dozens of specialized software vendors and consultancies have cropped up over the past two and half decades to address this; many top tier analyst firms have been putting out frameworks for almost three decades; and every Big X consultancy has their detailed methodology, 200 page strategy and implementation deck, and a dozen or so publications on their website documenting numerous benefits and case studies.

In fact, there’s so much material out there that if you asked your favourite clod or chat, j’ai pété instance to write you a book, put together an implementation plan, prepare an executive presentation, prepare an implementation plan for your team, and prepare a training plan for your supply base, chances are, believe it or not (and the author rarely does) any particular Gen-AI instance is likely to do a decent job of it because there is so much nearly identical material on the subject that the models are likely to be decently weighted to generate a decent facsimile of the material.

But here’s the question — is everything that has been defined, prescribed, optimized, and re-prescribed over the last two and a half decades the right approach? This might appear dumb on first read and sound dumber on first vocalization, but it’s actually pretty poignant. Just because whatever has been prescribed for the past two decades has worked just fine, that doesn’t mean it’s optimal. And just like modern drugs can deal with most symptoms of any ailment you might get, at least at first, that doesn’t mean they address the root cause. And if they don’t address the root cause, you can still get really sick, and even die if the symptoms the drugs are addressing are covering up a long term, slowly progressing, [incurable] cancer or autoimmune disease.

Plus, certain practices are designed to achieve end-states with certain assumptions in mind that are based on expected market conditions and activities that are expected to optimize to those conditions. In other words, its assumptions across the board (on what to do) based on assumptions across the board (on what is success) based on expectation across the board where the assumptions on what to do assumes that what got a result before will again (without any proper definition of the scenario in which it worked or proof that its effective) and the assumptions on what is success is only based on what has achieved the best results (and not the best results that could be obtained) and the expectations are simply hopes that what came before will come again.

Of course, given the reliability of some of the methodologies out there, the consistent results obtained by the best-in-class providers, and the wide-spread adoption of certain methodologies, it’s hard to argue with the assumption, the underlying assumptions, and the expectations because, for most organizations that adopt the methodologies, and stick to them, they get great results!

But what happens when the results flatline?

I’m asking because I expect there’s a chance it will happen, and possibly sooner than I expect, for companies who are already practicing good supplier performance management. There are a number of reasons for this. As per above:

  • effective is not necessarily optimal
  • you can get short-term results treating the symptoms, but you can’t get long term results unless you address the route cause
  • everything is based on assumptions, on assumptions, against expectations

and

  • market conditions are (drastically) changing
  • all good things must come to an end
  • managing after the fact is not the same as laying the groundwork for success

These are the reasons in particular I suspect there is a chance that current SPM frameworks and efforts might crash.

All Good Things Must Come To An End

This universal truth holds true for anything with a limit. Since every action has a limit when it comes to the minimum time required for completion, the maximum number of uses or lifespan of a product, the value that can be delivered against a metric, and so on.

Combine this with the definition of supplier performance management which focusses on measuring, analyzing, and managing vendor performance to cut costs, minimize risks, and drive continuous improvement, this means that SPM has a limit. This, of course, is nothing new — we’ve known this for decades. But, applied properly, SPM has worked spectacularly for decades because:

  • only so many suppliers can be measured, managed, and developed at a time … so there’s an endless supply of projects
  • suppliers constantly change … increasing the endless supply of projects
  • technology in our space has kept improving, allowing for slow, continual, improvement
  • the global trade situation was relatively static for two decades
  • most companies haven’t, and still aren’t, doing SPM in earnest
  • SPM is done by humans, who aren’t perfect and continuously make mistakes

It will continue to work spectacularly for the majority of organizations the majority of the time as a result of all of these reasons. But it may soon start sputtering for some organizations. The reasons being:

  • the suppliers not measured, managed, and developed are the ones that will need it the most and cause you the most problems
  • suppliers will suddenly become (effectively) unavailable unexpectedly — often before the contract expires
  • technology has stalled in SPM
  • the global trade dynamic has drastically shifted and has entered a new age of continual instability
  • those doing SPM are running on outdated assumptions
  • Agentic AI is being injected everywhere, and decreasing the reliability and effectivness

This means that the days might be numbered for SPM-in-a-box, which is pretty much what you get because most vendors / consultancies who do SPM just do SPM.

Which is concern #1.

More importantly, SPM is pointless if the supplier is going to suddenly become unavailable or unable to produce the products that you need.

A sanction, border closing (due to conflict or epidemic with pandemic potential), seaway closing, economic collapse, etc. can all take a supplier out of the picture overnight. Chances are you’re monitoring for this for your (most) strategic suppliers, can usually see it coming with enough warning to at least take some emergency measures.

However, chances are that this sub-tier supplier, or it’s sub-tier supplier, aren’t monitoring for potential disruptions like you are. What happens when those suppliers suddenly shut down or become unavailable. Then your supplier is unable to produce the products you depend on, and it’s essentially the same as that supplier suddenly becoming unavailable.

Which means you need Supply Chain Visibility in order to not only select appropriate suppliers, but ensure your suppliers are selecting appropriate sub-tier suppliers and, if not, help them to do so. However, not only are most companies still not adopting SCV solutions, but they aren’t part of standard SPM.

In addition, it’s not just availability, it’s supply line, and dependency. Rare earths are becoming more rare, oil keeps going up, and many renewable commodities in the food stuff categories are in relatively short supply due to population increase, droughts, and wildfires. Sufficient supply demands that dependency on any one raw material be minimized to the extent possible. This means that PLM, and especially design, alternate designs, and substitution capability is becoming more and more critical. But this is completely ignored by today’s SPM that just focusses on making the production process more efficient, not the product.

In other words, while the classic SPM theory is fine, and the implementation close to flawless in a handful of suppliers, the impact will soon be minimal to non-existent if they continue to address the wrong problem or try to address the right problem in a vacuum.

Now that process performance is often not the critical problem, the effectiveness of traditional SPM may begin to wane — fast.

Which brings us back to the starting question:

How Do You Execute Effective Supplier Performance Management in Today’s World?

I don’t have a good answer yet, but I believe you need to find one before the question becomes relevant.

Modern Strategies for Supply Chain Finance

The strategies for success haven’t changed much in the past few decades, but for some reason still aren’t in widespread practice. So we’re going to review the best practices from two decades ago and then indicate what else you should be doing.

1. Open Account and Letter of Credit Balancing

It’s critically important to understand your organization’s cost of capital versus your supplier’s cost of capital. Open account terms, for example, may bear lower fees than a letter-of-credit based transaction, but they can also restrict a seller’s access to working capital financing and increase its costs of working capital. The additional cost borne by the supplier for accepting extended payment terms, for example, could be finding their way back into your cost of goods sold.

2. Better Forecast Accuracy

One of the best ways to take cost out of the supply chain is to take unnecessary inventory out of the chain, as this just leads to additional storage, overhead, and financing costs and losses when it has to be cleared at considerable markdowns.

3. Supplier Cost Reduction

Depending on where your primary supply base is located, (up to) half of your suppliers might be struggling to access financing at acceptable terms. It might be their top issue, even above reliability of component and raw material supply! Their cost reduction is your cost reduction and your best chance at good supply chain finance strategies!

Building on these foundations, today you need to be considering:

4. Deep Tier Financing

Making sure your suppliers’ critical suppliers have access to fair and affordable financing so they can manufacture and sell their products affordably to your suppliers’ and not inflate your supplier’s cost which would inflate your cost.

5. Sub-Tier GPO

If you are relying extensively on a few key raw materials or components across your product lines, and if those are in short supply, costly, or volatile, buy the raw materials on behalf of your sub-tier supply base and sell it back at cost.

6. 0 Days DPO

Pay cash-strapped suppliers who would have to borrow at loan shark rates of interest immediately. It’s that simple. You’re paying for their overhead, so if your cost of finance is 1/5 or 1/10 of their’s, then, even if you risk borrowing occasionally, it just makes good financial sense.

7. Inventory Financing

Either enable your supplier to get financing based on their inventory materials and components and the goods they are going to produce, or create your own inventory financing solution for them.

The reality is that the standard techniques of JIT (and forcing suppliers to sit on inventory), increasing DPOs to the maximum legally allowed, and pretending EPD is a substitute for financing is NOT financing, only hurst the supplier, and only hurts you.