Most sales training treats the buying decision as the last event in a process: you did discovery, you built the business case, you handled objections, and now the customer decides. The evidence says the opposite. The decision is not an event at the end. It is a distributed, slow, socially negotiated process that runs underneath the entire deal, and the single most common outcome is that it never completes at all.

That last point deserves emphasis because it reframes what sellers are actually up against. Analysis of 2.5 million recorded B2B sales conversations behind Matthew Dixon and Ted McKenna's The JOLT Effect (2022) found that between 40% and 60% of qualified opportunities end in no decision. Your most frequent competitor is not the other vendor on the shortlist. It is the absence of a decision.

This article gives the Sales Agent a working definition of a B2B buying decision, the four distinct ways decisions fail, and what the research says actually moves each one.

A B2B buying decision is a group act of committing shared organizational resources under uncertainty. It completes only when enough members of the buying group have individually resolved enough of their own uncertainty to accept collective accountability for an outcome none of them can guarantee.

Three parts of that definition do real work.

Group. Forrester's The State of Business Buying, 2026 (published January 2026) found the typical business buying decision now involves 13 internal stakeholders plus 9 external influencers. That is not a committee you can meet. It is a network, most of which you will never speak to, and which deliberates without you in the room.

Under uncertainty. No buying group has enough information to be certain. They have enough information to be uncomfortable. This is the state sellers actually operate in, and it explains why "more information" is so often the wrong intervention.

Accept accountability. Someone signs. Someone's name is attached when the implementation slips or the promised savings do not appear. B2B decisions are made by people whose careers carry the consequences, and the research on why deals die keeps pointing back to that fact.

A useful mental model: a B2B decision has four locks on it. All four must open before commitment happens, and they open in no fixed order. A stalled deal is not stalled generally. It is stuck at a specific lock, and the intervention that opens one lock will do nothing for the others (and in some cases will make them worse).

The group must believe that not acting is worse than acting. Until then, doing nothing is free.

This is the oldest and best documented bias in the set. Samuelson and Zeckhauser's "Status Quo Bias in Decision Making" (Journal of Risk and Uncertainty, 1988) demonstrated across a series of controlled experiments and field data that people disproportionately stick with a current option simply because it is current, and that the bias strengthens as the number of alternatives grows. Kahneman and Tversky's prospect theory (Econometrica, 1979) supplies the mechanism: losses loom larger than equivalent gains, so a change that risks a known state is weighted more heavily than its arithmetic upside justifies.

There is a practical corollary that most sellers already sense: Gartner's B2B buying journey research reports that 99% of B2B purchases are driven by some organizational change. Absent a trigger (new leadership, a merger, a regulatory shift, a failed renewal, a growth target that current tooling cannot support), the status quo usually wins on its own.

The JOLT research puts a number on this lock: 44% of no-decision losses trace to a customer preference for the status quo.

The group must be able to tell the options apart and rank them. Where Lock 1 is about whether to act, Lock 2 is about whether they can see clearly enough to choose.

This is where the evidence gets more interesting than the folk wisdom. Chernev, Böckenholt and Goodman's meta-analysis of choice overload (Journal of Consumer Psychology, 2015), covering 99 observations from 53 published studies and 7,202 participants, found that the mean effect of assortment size on choice overload is not significant on its own. More options do not automatically paralyze people. The effect appears reliably only under four moderators: high decision task difficulty, high choice set complexity, high preference uncertainty, and an effort-minimizing decision goal.

Read that list again, because it describes an enterprise software evaluation almost perfectly. Time constrained, accountable, many attributes, options that are hard to compare directly, buyers who do not have articulated preferences yet, and a committee trying to reach a defensible answer without unlimited effort. B2B buying is the exact condition under which choice overload does bite.

Gartner's 2019 study of more than 1,000 B2B customers found the same pattern from the supplier side: customers who experienced information overload were 153% more likely to settle for a smaller, less disruptive course of action than they originally planned. Note what happens there. Overload does not usually produce a loss. It produces a shrunken deal, which shows up in your pipeline as a win.

Each individual must be able to live with being wrong.

This is the lock most sellers underweight, and it is the largest single driver of no-decision. In the 2.5 million call corpus behind The JOLT Effect, 56% of no-decision losses were attributable not to status quo preference but to customer indecision: the fear of making a mistake with this purchase, this vendor, this timing. Dixon and McKenna found moderate or high levels of indecision present in 87% of opportunities. Fear of messing up is close to a universal condition in the pipeline, not an edge case.

The most important finding for practice is what happens when sellers misdiagnose Lock 3 as Lock 1. The standard reflex when a deal stalls is to intensify the cost-of-inaction argument. In the JOLT data, doubling down on status quo pressure with an indecisive customer backfired 84% of the time. This makes sense once the mechanism is clear. A customer who is afraid of choosing wrong, when told that not choosing is dangerous, is now afraid in two directions at once. The response to compounded fear is paralysis, not action.

The interventions that did work were different in kind: giving a firm recommendation rather than presenting neutral options, actively limiting exploration instead of feeding more information, and de-risking the decision through mechanisms like phased rollouts, opt-outs, pilots, and trials that shrink the size of the bet. Forrester's 2026 data shows buyers reaching for these mechanisms themselves: over 60% of business buyers now use trials to evaluate solutions, rising to 78% on purchases above $10 million.

Enough of the group must arrive at the same answer at the same time.

With 13 internal stakeholders and 9 external influencers, this is a coordination problem, and it is not solved by convincing each person individually. Notably, Forrester found that 94% of buyers in groups of six or more report clear benefits from that group structure, so the buying group is not a bug the seller should try to route around. It is how these organizations have decided to manage risk, and it is growing rather than shrinking. Procurement has become a decision maker (not merely a process gate) in 53% of business buying cycles.

The failure mode at Lock 4 is subtle: a deal can have full problem conviction, clear options, and safe individuals, and still die because the group never converges. Different members hold different mental models of the problem, and nobody has reconciled them.

If the four locks describe the failure modes, the strongest evidence for what actually opens them comes from Gartner's research on what it calls sense making.

Gartner's 2019 study of more than 1,000 B2B customers compared three seller approaches to information. Giving information (send more content, more case studies, more data) and telling information (assert your point of view, push a prescriptive answer) both underperformed. The third approach, sense making, produced high-quality, low-regret deals for 80% of the sellers who used it. Sense making means helping the customer evaluate the information they already have: connecting them to relevant sources, clarifying and filtering the complexity they have collected, and collaborating on how to judge quality rather than telling them what to conclude.

The mechanism appears to be decision confidence. A separate Gartner study of over 1,000 B2B customers (2019) found that customers confident in their ability to navigate a buying decision were 2.6 times more likely on average to expand an existing relationship. A later Gartner survey of nearly 1,000 B2B customers (November to December 2020) found customers with high decision confidence were 10 times more likely to make a high-quality, low-regret purchase.

Two related findings should settle a live debate about whether sellers still add value. Gartner reports that 43% of B2B customers would prefer not to interact with a sales rep at all, yet customers who avoid reps report 23% higher purchase regret, and buyers who use supplier digital tools in partnership with a rep are 1.8 times more likely to complete a high-quality deal. The preference for a rep-free experience is real. It is also, on this evidence, not in the buyer's interest. The seller's job is not to insert themselves into the journey but to be the thing that makes the buyer's own information usable.

Diagnose the lock before choosing the intervention. When a deal goes quiet, the question is not "how do I create urgency." It is "which lock is closed." Weak trigger event and comfortable incumbent points to Lock 1. Endless comparison, expanding requirements, and new evaluation criteria appearing late points to Lock 2. Repeated requests for more references, more proof, more security review, and a champion who agrees with everything but will not schedule the next step points to Lock 3. Alignment in every individual conversation but no movement in the group points to Lock 4.

Stop treating information volume as a service. More content is a Lock 2 accelerant. The evidence favors curation, an explicit point of view about what matters and what does not, and helping the buyer judge the quality of what they already have.

Give a recommendation. Neutrality reads as abdication to a buyer who is afraid. The JOLT finding on firm recommendations is one of the more actionable results in modern sales research, and it runs against the consultative instinct to present balanced options.

Sell the size of the bet, not just the size of the prize. Pilots, phased scopes, opt-out clauses, and success-gated expansions are Lock 3 instruments. They work because they change the personal downside for the individual signing, which is the actual variable in play.

Instrument for consensus, not just for the champion. Multithreading is not a coverage exercise, it is a Lock 4 exercise. The question is not "have I met the CFO" but "do the CFO and the VP of Operations describe the problem the same way."

Watch for the shrunken win. A deal that closes at half the intended scope is often an overload symptom, not a budget one. Gartner's 153% finding suggests these are the most invisible losses in any pipeline, because they look like revenue.

The through line across every study cited here is the same. B2B buyers are not short on information, options, or vendors willing to pitch them. They are short on confidence. The seller who reliably increases decision confidence, across a group, without adding to the pile, is doing the only part of this job that the evidence says changes outcomes.