# VML study finds 38% of enterprise B2B purchases end with no decision

By MarketScale Newsroom · Published 2026-11-14 · Marketing Tech on MarketScale
Canonical: https://www.marketscale.com/industries/marketing-tech/vml-study-finds-38-of-enterprise-b2b-purchases-end-with-no-decision

> VML surveyed 750 B2B buyers: AI starts most supplier searches, but finance-led committees and career risk leave 38% of big purchases undecided.

## Key points

- The supplier that wins the AI search can still lose the approval meeting: 71% of B2B buyers pick the safe, well-known name over a rival they believe offers better value.
- Risk and compliance (39%) and budget or ROI uncertainty (34%) top the stall reasons, suggesting sellers should put ROI, privacy and integration proof in front of finance and risk reviewers early—before late-stage internal alignment drags on.
- A benchmark for any buying committee: an average of 6.8 stakeholders and 5.35 extra weeks of alignment after a preferred supplier has already been picked.

Enterprise buyers face a lot of pressure when they set out to make a big purchase. VML's Future Buyer Report 2026 surveyed 750 B2B buyers across eight international markets to understand how they approach significant procurement initiatives and what weighs on their decisions.

The study, led by Dr. Andreas Ortner, VML's managing director for B2B in EMEA, covers seven sectors: manufacturing, energy, transportation, finance, professional services, telecommunications, and healthcare and beauty. It extends VML's long-running Future Shopper consumer research, which a 4As webinar listing describes as drawing on 28,000 consumers across 17 markets. The B2B sample is far smaller, which is worth keeping in mind when reading the sector splits.

Most coverage will grab the finding that generative AI has overtaken Google as the place supplier research begins. For anyone who runs a buying process, the useful part sits between the two ends of the report. AI is speeding up how a shortlist gets built. A finance-led committee, conscious of careers, then slows how that shortlist becomes a signature, and that gap is where the 38% goes.

VML's launch video reduces its argument to three tests. AI has to surface a supplier's brand, people have to trust it, and the buying group has to sign off on it. Most of the reasons buyers give for stalled deals fall at that last approval step, which suggests the 38% mostly reflects a failure on the third test rather than the first.

## The shortlist forms before the sales call

In VML's data, 55% of buyers begin supplier research on a generative AI platform, against 45% who start with Google or Bing. The ordering matters because of what happens next. Brand discovery is happening before human outreach occurs, and 74% of buyers say they are likely to act on AI-generated recommendations.

Block Field

VML's reading is that traditional relationship-led pitching has lost influence in early discovery. A buyer who has already formed a preference before the first call is not waiting to be persuaded; they are checking whether the supplier matches what the research told them.

The manufacturing cut of the report says AI now lets buyers in that sector shortlist suppliers earlier than before, and the transportation cut lists boosting search and AI visibility among the moves suppliers need to win deals. VML's prescription for suppliers is generative engine optimization: publishing product specifications, ROI data and API integration credentials in a structure AI crawlers can parse and recommend.

That makes it a documentation job as much as a marketing one. Spec sheets, integration pages and pricing explanations become the raw material a model quotes back to a buyer. The hiring market is starting to reflect the shift. The American Marketing Association's 2026 State of Marketing Careers Report, citing an analysis of global job postings, lists GEO/AEO specialists among the marketing roles on the rise, while SEO specialist roles fell 15%, the steepest decline of any role tracked. That suggests making a company legible to a model is becoming a job in its own right.

Procurement teams have a stake in this too. If an AI tool assembles the first list, that list reflects what suppliers chose to publish in machine-readable form, which could favor the best-documented vendors over the best-fitting ones. Organizations where staff lean on AI for early research could reasonably ask which tools are in use and who checks the first list before it hardens into the shortlist.

## Finance holds the pen

Who signs is a different matter. Asked who holds the greatest influence over final supplier approval, buyers put finance first and the people who usually feel the need last. Business unit leaders carry meaningful sway in only 40% of major decisions; front-line sales and marketing barely register.

Block Field

A significant purchase now involves an average of 6.8 stakeholders. VML also found it takes an additional 5.35 weeks of internal alignment after the group has already identified a preferred supplier, more than a month spent selling a decision that, on paper, is already made. The gatekeeper shifts by sector: IT has the most influence over telecommunications procurement, while internal finance gatekeepers govern buying in financial services.

VML says purchasing power is moving away from front-line business unit leaders and consolidating with centralized corporate gatekeepers. Its warning to sellers is pointed: a pitch aimed only at the departmental end user gives them a false sense of security. VML recommends multi-threaded sales collateral, with separate material on financial ROI, data privacy and systems integration, so the case holds up under centralized committee review. The telecom split shows why a single deck rarely covers it. Where IT leads approval, the integration document could count for as much as the ROI model.

Then there is the personal arithmetic. Eighty percent of buyers say they consider how a purchase will affect their professional reputation, and 73% agree their career would take meaningful damage if a purchase underperformed. VML calls the result defensibility bias, and this is where its sharpest number lands.

The supplier that wins the AI search can still lose the approval meeting: 71% of buyers pick the safe, well-known name over a rival they believe offers better value.

That helps explain why brand reputation, cited by 49% as a top selection factor, sits level with provable ROI, which 49% rank as their leading investment criterion. Reputation is also the most common tiebreaker when buyers narrow a shortlist, at 30%. It also explains why 65% of buyers call third-party analyst recognition essential or important.

VML's view is that challenger brands cannot win on price or feature superiority alone. For a challenger supplier, the evidence suggests the person who found them is the one who needs ammunition. VML recommends handing that internal champion board-level risk mitigations, analyst validation and verified customer case studies. Its professional services cut frames the same task as earning trust that carries through to approval.

## Where the 38% goes

Risk and compliance concerns are the most-cited reason initiatives end without a decision, at 39%, followed by budget or ROI uncertainty at 34%. Process friction follows: shifts in corporate strategy (31%), procurement complexity (27%) and misalignment inside the buying group (26%). The figures add up to well over 100%, so buyers evidently named more than one cause.

Abandonment runs highest in telecommunications, at 44.1%, and finance, at 40.8%. Energy has its own version of the problem: VML's summary of its energy sector report says 52% of energy buyers report that unclear ROI can stall a major initiative.

Read against the approval chart, the top two stall reasons, risk and money, are exactly the questions the most powerful approvers are positioned to ask. That suggests many stalls happen when evidence reaches finance and risk reviewers late, or in the wrong form. VML puts it more bluntly: in enterprise sales, the real competitor is often internal inertia rather than a rival supplier.

Not every stalled deal is waste, though. With 31% of buyers citing a shift in corporate strategy, part of the 38% reflects purchases that stopped fitting the plan and probably should have stopped. The figure works better as a ceiling on avoidable stalls than a measure of them. The portion a procurement director can actually control is the friction: complexity, misalignment and proof that shows up after the reviewers have already formed a view.

Block Field

Suppliers can meet that halfway. Transparent pricing models are the factor buyers most often say make a supplier easy to buy from, at 39%, and VML pairs that with clear, measurable implementation roadmaps as the way to take risk off the table early.

## People still settle the final round

Buyers are not asking software to finish the job. Some 64% of B2B transactions happen digitally, VML found, yet only 3% of buyers want a purely digital experience. They want people for commercial negotiation (47%), escalating problems after purchase (43%) and evaluating complex or customized solutions (42%). The healthcare and beauty report makes the same point plainly: buyers there use AI to assist major investment decisions, not to lead them.

The tiebreakers follow the same pattern. If two suppliers are comparable on function, the factor that most often settles it is responsiveness and how easy the team is to work with, at 22%. Brand reputation follows at 18% and price at 14%. That order suggests that in a close contest, the service team could count for more than a well-known name.

Set that beside the shortlist stage, where reputation led as a tiebreaker at 30%, and a sequence emerges. A known name helps a supplier get onto the list. Once the rivals look alike on paper, how the team behaves carries more weight than the logo.

VML's prescription for sales follows from that. It argues the B2B salesperson's role has shifted from pitching to consultative problem-solving, with sales teams acting as technical and commercial advisors who guide buyers through risk reviews, integration scoping and contract negotiation. Those are the same pressure points behind the top stall reasons, which could make that late-stage advisory work one of the few levers a supplier has over the 38%.

The split shows up well outside enterprise buying. Zillow's Consumer Housing Trends Report 2026, written by Edward Berchick, found 82% of home buyers use AI in general and 61% of those users used it while buying a home, yet 85% of buyers still worked with an agent. The most common uses were learning about the process (74%), getting market insights (71%) and finding homes (70%), all research jobs. Different market, different stakes, same division of labor: AI does the research, a person handles the deal.

Ortner sums up the study this way: the B2B supplier of the future has to be ready for procurement and has to pair discoverability in AI with the trust of people. The figure that could upset that balance is VML's 31%, the share of buyers who expect an AI agent to be the primary evaluation mechanism for their next major purchase. If that happens at scale, evaluation shifts into software while approval stays with a finance team and a C-suite worried about their reputations. The next round of buying decisions will show whether those reviewers sign off on a supplier an agent picked.

## Sources

- [The Future Buyer Report 2026](https://www.vml.com/insight/the-future-buyer-report-b2b-2026)
- [The Future Shopper 2026. Winning Tomorrow's Consumer](https://www.aaaa.org/course/the-future-shopper-2026-winning-tomorrows-consumer/)
- [Buyers: Results from the Consumer Housing Trends ...](https://www.zillow.com/research/buyers-housing-trends-report-2026/)

Tags: VML, Future Buyer Report 2026, B2B buying, procurement, generative AI, generative engine optimization, enterprise sales, buying committees, supplier selection, finance approvals, sales operations, B2B marketing

---
Source: MarketScale, https://www.marketscale.com/industries/marketing-tech/vml-study-finds-38-of-enterprise-b2b-purchases-end-with-no-decision. Published for AI indexing and citation; cite the canonical URL. Site guide for agents: https://www.marketscale.com/llms.txt