two numbers that moved in opposite directions

SAP published the fifth consecutive year of the Economist Enterprise Procurement Imperative research on September 17, and the five-year view is more interesting than any single finding in it. The 2026 edition is titled "Procurement at a Crossroads: From Optimism to Realism" and surveyed 2,648 C-suite executives across 23 countries. Gordon Donovan, global vice president of Research, Procurement, and External Workforce at SAP, wrote the summary.

Start with collaboration. In 2023, 53% of executives agreed that procurement collaborated effectively with the rest of the organization. That rose to 75% in 2024 and 90% in 2025. SAP publishes no 2026 collaboration figure, so 90% is where the series stops, and it describes last year rather than this one.

Now the other line. By 2026, executive confidence in procurement's influence over digital transformation had fallen to 68%, down from 91% in 2025. In the same survey, 60% name digital transformation as procurement's top strategic priority for the next 12 to 18 months, up from 38% a year earlier. Procurement has never wanted the technology agenda more, and the people above it have never been less sure procurement should be the one making those calls.

what that gap usually means inside a company

A rising collaboration number and a falling influence number often describe the same change from two angles. Procurement gets pulled into projects earlier. Its spend numbers get quoted in more meetings. And then the platform decision gets made somewhere else, by IT or by a transformation office, with procurement listed as a stakeholder rather than an owner.

The tell is usually in who holds the implementation budget and who owns the integration backlog. A function can sit in every requirements workshop and still have no say in which system gets bought, because the money and the delivery capacity live in another org. Anyone who has watched an integration map get drawn after the contract was signed knows how that ends.

The survey adds one structural detail that fits. By 2026, 44% of procurement leaders reported to the COO, up from 33.5% in 2022. Reporting into operations puts procurement closer to execution and further from the budget line that funds enterprise software. SAP's page does not connect those two findings, and the connection is our read rather than the research's.

the barrier moved from money to practicality

The most useful number in the set might be the one about obstacles. Ease of deployment is now the primary concern for 71.7% of respondents, against the 30.2% who named budget back in 2022. What stands between procurement and the tool it wants stopped being approval and started being delivery.

That changes what a procurement leader has to bring to the steering committee. A budget objection is answered with a business case and a payback period. A deployment objection is answered with data readiness and a named owner for the integration work, plus an honest estimate of how much of that work lands on a team already booked through the year. The questions in a vendor demo you actually control get sharper when ease of deployment is the stated blocker, because a demo environment is exactly where deployment difficulty hides.

Two more findings sit uneasily next to each other. AI-driven and predictive insight is the dominant category-management priority for 66.6% of respondents, while 54% still identify cost savings as procurement's primary contribution. A function measured on savings and asked to fund predictive tooling ends up making the same argument every budget cycle, the same squeeze the clean core budget argument describes on the ERP side.

who paid for the research

SAP sponsored this research, SAP published the summary, and the author is an SAP vice president. None of that makes the numbers wrong. It does shape which findings get a headline, and a sponsored survey concluding that ease of deployment is the top barrier points neatly at a category of software the sponsor sells.

The missing 2026 collaboration figure is the other thing we would flag. The page walks the series from 53% to 75% to 90% and then stops. Whether the question was asked this year and left unpublished, or not asked at all, the post does not say, and that is the part we would ask about. The 90% belongs to 2025, and nothing on the page makes it current.

This desk covered the same research programme two days ago, through the Ariba spend analysis announcement, where the confidence number arrived inside a product launch. The five-year series is the more useful artifact. One year of movement can be noise. Three years of climbing collaboration scores followed by a one-year collapse in influence is a shape somebody should explain.

what to check before your next steering meeting

Pull the last two technology decisions that touched procurement in your organization and answer four questions about each. Who wrote the requirements. Who signed the contract. Who owned the deployment plan. Who got the call when the rollout slipped.

If procurement wrote the requirements and someone else owned everything after that, the 68% describes your organization too, and no amount of collaboration credit fixes it. That calls for a named deployment owner inside procurement with budget attached, the same discipline a readiness check before the first real rollout asks for.

So here is the question for the next steering meeting. When the next procurement platform gets bought, who signs the statement of work, and who owns the date it goes live? If the room needs more than a few seconds to answer, the influence number is already telling you something about your own organization, whoever sponsored the survey that produced it.