About a month ago, Jonathan O’Brien of Positive Purchasing and Guy Strafford of OneSupplyPlanet released a white paper on this very topic where they claimed:

AI means Procurement is going to be left jostling with the rest of the business to control the commercial space. It is a race. Start now or lose.

I have to say I don’t agree. In fact, we’re still in the start now AND lose timeframe, as evidenced by the recent MIT study demonstrating that 95% of AI projects have failed and the plain and simple fact that the majority of consultancies, providers, and new-age services-as-software providers claiming to provide AI Employees (which is all bullcr@p by the way, see our previous posts as AI Employees Aren’t Real) are applying the wrong AI to the problems they believe AI can solve. (And that’s another issue, real AI can do quite a bit, but not what the greatest con-man since PT Barnum not in politics keeps telling us it can do.)

However, they’re close. You need to start learning what AI is and isn’t now, looking out for the past and next (NOT current) generation of providers who were applying real AI properly and who will emerge to apply next-gen implementations of real (non-LLM) AI properly to Procurement problems (with real education and real procurement experience to back it up, and that’s not a Youtube crash course in how to engineer a prompt for a bullcr@p LLM, by the way). That way, you can select the right solution for the right problem at the right time.

You are going to have to jostle with the rest of the business, especially Supply Chain who believes they are the most critical function (when they are only one side of the critical coin with Procurement being the other half), IT who believes they have the best technical knowledge and should make the decision (and don’t really understand what the tech has to do for the rest of the business, just what is easy for them to maintain and fun for them to play with), Finance who wants to ensure they have better visibility and control over every dollar, Sales and Marketing who want to pretend the age of the Mad Men never ended, and so on.

Moreover, while many of the activities carried out by Procurement functions will be automated, these are just the data entry, transformation, and output functions that we were promised would be automated 40 years ago when computers started entering the average business. We’re not going to see the automation of the strategic functions, the relationship management, or the consensus building that is critical to success. While the authors may claim that many Procurement activities will soon be managed directly by the business, that will be a step back as we’ve seen, and Hackett has catalogued, the significant difference in performance in businesses that have standalone best-in-class Procurement departments and those that don’t have any Procurement departments. The reality is that, in the economic and technological climate that is coming, businesses without Procurement likely don’t have the same chance of survival as businesses with Procurement.

According to the authors, all that will remain are the commercial hub activities that require human expertise and intervention: change management, innovation, and sophisticated market engagement. While these will definitely remain, the reality is that you can’t turn supplier discovery, qualification, onboarding, relationships, performance management, development, and innovation over to a machine. You can’t build organizational consensus through emotionless algorithms. Not all award decisions can come down to the results of lowest cost computations after automated negotiations based on bid rank or classic game theory (or even modern game theory as many of the game theory “experts” get this wrong regularly, as the author has poked holes in claimed “optimal” solutions presented on LinkedIn and vendor websites more than once, because there’s always an assumption as to what optimal is, and it’s usually one sided or wrong). Technology will reduce the time requirements for a lot of these processes as it will fully automate the data collection, transformation, analysis, and recommendations for you, as a human expert, to one-click accept or deny, but a human will still be needed. All that will disappear is the 80% to 90% of the work that is tactical data processing. While this will displace people, as all technological evolutions do, we need to remember that each evolution has ultimately resulted in the creation of new jobs as old ones get automated. Not only will people be needed to maintain the automations and hardware supporting them, but new strategic and creative jobs, some of which we can’t yet predict, will emerge as a result.

Moreover, the authors believe that since such capabilities are found across the business, the other functions … will want to move into this commercial space and that unless Procurement develops the skill sets to a higher standard than the other functions, it will be outpaced.

While they are not wrong, and while this will make life difficult for Procurement if other functions get ahead of them in terms of value delivery to the business, Procurement is more than just change management, innovation, and market engagement. However, without this core, Procurement’s differentiation will be limited and its overall influence over the business not what it should be.

Moreover, while there will be a battle to evolve and survive post AI, we’re not there yet because what we have now is not AI, it’s the latest instantiation of Silicon Snake Oil with grandiose, false, claims and no real value. We’re still a few years away from widespread application of real, useful, AI, and more than a few years away from post-AI. In other words, the likelihood of Procurement being phased out as-is by 2035, as the author’s claim, is not too likely at this point. However, depending on how fast we get to, and through, the Fake AI crash and on to real AI, Procurement could be in deep trouble in the early 2040s. Which is why you have to start learning about real AI today, where you can apply it safely and effectively, and how you can implement it bit by bit for stable, guaranteed, success. Learning — not rushing in to an incomplete, half baked, solution guaranteed to make you the next casualty among the 95% error rate — is key.

So for those of you who asked, that’s my initial response as to whether or not there is a Functional ExtAInction Battle in Procurement. There’s the same battle there’s always been, but without real AI, and without the rest of the business having the deep Procurement knowledge necessary for real Procurement success (which goes well beyond what can be automated today), there’s no chance of ExtAInction in any Procurement department which is a leader in its operation.

However, even though the conclusion is wrong, the majority of the observations and analysis in the paper is right. In fact, it’s one of the best the doctor has read yet in terms of analysis and insight. So in our next post(s) we are going to discuss that. Despite a few mistaken conclusions, which can be forgiven because it’s really hard to understand the reality when it takes a very advanced understanding of mathematics to understand the tech, which is necessary to understand the reality, it does a great job of figuring out how Procurement needs to be seen and why. So download the white paper and read it today for the insights within (without having to worry about an extinction that won’t happen … at least not yet).