Why a Report Can't Tell You What to Do Next

Spend analytics are not the problem. A function that knows where it spends, which suppliers are outliers against benchmark, where contracted savings failed to land in the P&L, and which categories have the most concentrated risk is in a better position than one that does not know these things. The investment in data infrastructure, dashboarding, and category intelligence is a genuine one and its outputs are genuinely useful. The question is what they are useful for, and what they cannot do, and the answer to the second part is where most analytics programmes quietly run out of road.

A report is retrospective by design. It explains what happened. It identifies where the gap between contracted price and actual spend opened up, which negotiations produced savings that did not arrive, which suppliers have been extracting rent through auto-renewal clauses nobody reviewed. The insight is real. The problem is that insight and the behaviour change that would close the gap it identifies are in different parts of the stack, and data infrastructure reaches only one of them.

What analytics is genuinely good for

Before naming the ceiling it is worth naming the floor, because the floor is high and the work is real.

A well-built analytics function can tell a procurement leader which ten suppliers represent eighty percent of the function's commercial exposure, where category spend has drifted outside contracted terms, how the function's achieved prices compare to market indices, and where the savings claimed in negotiations matched the savings that eventually reduced actual cost. Each of those answers is hard to get without good data and represents genuinely important work that changes what the function prioritises.

Category strategy, supply risk analysis, supplier segmentation, and savings tracking all depend on the retrospective view that analytics provides. These are not small things. A function without this capability is flying blind in ways that a well-instrumented one is not.

Where the report ends and the gap begins

The limit appears at the edge of what retrospective data can reach. A dashboard that tells you Supplier A has been charging fifteen percent above benchmark for the past three years has identified an opportunity. It has not done anything to close it. Closing it requires having a conversation with Supplier A that is commercially credible, well-timed, anchored correctly, and held by a buyer who can maintain a position under pressure. None of those things are in the report.

This is not a design flaw in the analytics. It is a structural characteristic of what retrospective measurement can do. A report describes the past state of a system. The commercial outcome of the next negotiation depends on what happens in the next negotiation, which is a forward-looking behavioural event that no amount of historical data can run for you.

The gap shows up in a recognisable pattern. A procurement function invests in analytics, surfaces real savings opportunities, builds a category strategy that identifies exactly which relationships are underperforming commercially, and then those opportunities fail to convert into outcomes at the rate the analysis suggested they should. The usual diagnosis is that the data was wrong or incomplete. Sometimes it is. More often the data was fine and the gap is in the execution, in the actual conversations with suppliers, which the analytics pointed toward but could not improve.

Closing the loop

The phrase "data-driven procurement" usually means using data to identify what to work on. It rarely means using anything to build the capability to execute on what the data found. Those are different investments that get conflated, and the conflation is why functions can have excellent analytics and mediocre commercial outcomes simultaneously.

Closing the loop means connecting the insight to the behaviour that would capture it. That connection requires practice. A buyer who understands from the data that Supplier A is overpriced by fifteen percent still needs to be able to hold the position when Supplier A's account manager explains at length why the price is justified and the relationship would suffer if it were challenged. The data gives the buyer a reason to push. The skill determines whether the push lands.

The deliberate-practice piece in this series covers how that skill gets built rather than assumed. The practical point here is simpler: a measurement system that tells you what went wrong cannot by itself change what goes right next time. The insight that a report surfaces and the capability needed to act on it are genuinely different things, and a function that has invested in one but not the other has done half the work. Voice2Evolve exists for the other half: turning the opportunities the data identifies into the conversations that capture them.