Qualification is the stage where sellers decide where their time goes. Everything else in a sales process is execution. Qualification is allocation, and allocation is the decision that compounds: a quarter spent on the wrong six opportunities cannot be recovered by better discovery calls, sharper demos, or more disciplined follow-up on the remaining four.

Most sales organizations treat qualification as a form to complete. A seller fills in fields, a manager reviews them in a pipeline meeting, and the deal advances or does not. The evidence suggests this ritual is close to useless, and occasionally worse than useless, because it produces the feeling of rigor without the substance of it. The problem is not that qualification frameworks are wrong. It is that the most popular ones answer a question that is not the one killing deals.

Start with the outcome data. In the research behind The JOLT Effect, Matthew Dixon and Ted McKenna analyzed 2.5 million recorded B2B sales conversations and found that between 40% and 60% of deals that reach late stages end in no decision. Not lost to a competitor. Not lost on price. Lost to nothing at all. Across that call corpus, 87% of customers showed moderate or high levels of indecision.

Sit with the implication. If roughly half your losses are to no decision, then the modal failure in your pipeline is not "we were beaten." It is "the purchase never happened." And yet the qualification frameworks in widest use, BANT above all, are built almost entirely to assess whether you will win a purchase that is assumed to be happening. Budget, Authority, Need, and Timeline are all questions about the shape of a transaction whose existence is taken for granted.

MEDDIC and its descendants improve on this substantially. Identifying an economic buyer and a champion, and insisting on quantified metrics and a documented decision process, forces at least some contact with the question of whether the organization can actually get a purchase through. But in practice MEDDIC is usually deployed as a scoring rubric, and scoring rubrics have a well known failure mode: sellers fill them in from inference rather than evidence, and the score becomes a summary of the seller's optimism rather than a measurement of the deal.

There is a second problem, and the academic literature is blunt about it. Mayberry, Boles, and Donthu, writing in the Journal of the Academy of Marketing Science in 2018, studied effort allocation in the industrial salesforce of a large multinational and found clear escalation of commitment: salespeople "continue to commit resources to opportunities with little or no chance of being won, to the exclusion of viable leads." The effect was strongest among lower performing sellers and in non-strategic accounts. Their conclusion is uncomfortable and worth quoting: without targeted managerial intervention, sellers will over-invest in unwinnable opportunities. Compensation design does not fix it. Sellers do not walk away on their own.

So qualification has to do two jobs that most implementations do badly. It has to test whether a purchase will happen at all, and it has to create enough external pressure that a seller will actually abandon a deal that fails the test.

Qualification is the ongoing production of evidence about three separate propositions, each of which can be false independently, and each of which decays over time.

The three propositions are:

One: a purchase will happen. This organization will buy something, from someone, inside a defined window, and the internal machinery to make that happen (funding, approval, sponsorship, sequencing against other priorities) either exists or has a credible path to existing.

Two: the purchase will be ours. Given that a purchase happens, we are the choice, on grounds we can name, against alternatives we can name, held by people we have actually spoken to.

Three: the pursuit is worth its cost. The expected value of this opportunity, discounted by the probability of the first two propositions and by the time required, beats the next best use of the same hours.

Almost every qualification failure is a collapse of these three into one undifferentiated feeling about whether the deal is good. Kept separate, they behave very differently. A deal can be highly likely to be won and highly unlikely to happen: that is the classic late stage no decision, where you are the clear favorite in a race that gets cancelled. A deal can be certain to happen and unwinnable: an incumbent renewal dressed up as a competitive evaluation. A deal can pass both tests and still fail the third, because it will consume four months of a seller's capacity to produce a contract worth a fraction of what those months could have produced elsewhere.

The mechanism that makes this work is a discipline about what counts as an answer. Every qualification claim should carry four things: the claim itself, the person who is the source, the date it was established, and the observation that would falsify it.

"They have budget" is not a qualification. "The VP of Operations said on 14 August that the automation line item is funded at roughly 400k in the current fiscal year, and would be falsified if finance re-baselines in the October planning cycle" is a qualification. It has a source, a date, and a stated way to be wrong.

This sounds bureaucratic. It is actually the opposite, because it eliminates most of what fills pipeline reviews. A claim with no named source is a guess. A claim with a source but no date is probably stale: buying groups reorganize, sponsors leave, budgets get re-cut. A claim with no falsifier cannot be tested, which means it will never be revised, which means it will sit in the CRM justifying continued investment until the deal dies.

The decay point deserves emphasis. Qualification is not a gate you pass through once. It is a set of claims with expiry dates. Forrester's 2026 buyer insights research describes buying groups of roughly 13 internal stakeholders and 9 external influencers on a typical purchase, with procurement acting as a decision maker in 53% of buying cycles. A group that large does not hold still. Anything you established two months ago about who decides and on what basis should be assumed suspect until re-verified.

If the dominant loss mode is no decision, the practical question is how you test for it early, when abandoning costs you little.

The most direct evidence concerns money. Gong Labs analyzed 11,331 opportunities with at least three calls each and found that win rates were highest when pricing was discussed on the first call (42%, against 32% on the second, 15% on the third, and 5% when pricing was never raised). Budget showed a sharper pattern still: 49% when discussed on the first call, 7% when never mentioned. Notably, raising budget early did not shorten relationships, it lengthened them, roughly doubling the number of buyer and seller interactions.

The naive reading is that talking about price early causes wins. The more defensible reading is that a buyer willing to have a concrete money conversation on the first call is a buyer for whom a purchase is real, and a buyer who deflects money for three calls is usually not deflecting out of negotiating strategy but out of the absence of a live purchase. Early money conversation is a diagnostic instrument, not a persuasion tactic. Treat it that way and it does its work.

The second instrument is commitment to a next step. In a separate Gong analysis of 8,382 deals, opportunities where a concrete next step was established closed at 20%, against 5% where none was. Again the causal story matters less than the diagnostic one. A buyer who will not commit their own calendar to the next action is telling you something about proposition one, and they are telling you cheaply, early, and clearly.

The third instrument is group consensus. Gartner surveyed 632 B2B buyers in August and September 2024 and found that 74% of buying teams show unhealthy conflict during the decision process, with conflicting objectives or members overruled by external decision makers. Groups that reached consensus were 2.5 times more likely to report a high quality deal. The same research found something counterintuitive that bears directly on how sellers qualify: content tailored to individual stakeholder priorities reduced consensus by 59%, while content tailored to the group as a whole increased it by 20%. Qualifying a deal by collecting individual enthusiasm from stakeholders one at a time can therefore produce a confident forecast on a deal that is quietly fracturing.

None of this helps if the seller will not act on it, and the escalation of commitment research says they usually will not. Mayberry and colleagues found the tendency strongest among lower performing sellers, which is exactly the population whose pipelines most need pruning, and exactly the population least likely to prune them voluntarily.

Three structural fixes follow from the evidence rather than from folklore.

First, make the falsifier the object of review. Pipeline meetings that ask "what is the status" invite narrative. Meetings that ask "what would have to be true for this to be dead, and have you checked" invite evidence. The question is answerable, and it is answerable badly, which is the point.

Second, separate the verdicts explicitly in whatever system you use. A single 1 to 10 deal score lets a strong proposition two mask a weak proposition one. Three separate judgments do not blend.

Third, treat disqualification as output, not failure. The Mayberry finding is that sellers neglect viable leads in order to keep feeding unwinnable ones, which means every deal killed is capacity returned, not effort wasted. If your organization counts opportunities created but not opportunities correctly abandoned, you have built an incentive to escalate.

There is a related organizational failure worth naming. Sabnis, Chatterjee, Grewal, and Lilien, studying 461 sales representatives across four companies for the Journal of Marketing in 2013, documented what they called the sales lead black hole, in which roughly 70% of marketing generated leads receive no sales follow-up at all. Their finding was that lead prequalification quality and managerial tracking both drive follow-up, and that experienced reps in particular become more responsive to prequalification quality and less responsive to being tracked. Qualification, in other words, is not only a filter that removes bad opportunities. Done credibly, it is the thing that earns a seller's attention for the good ones.

Run the money conversation and the next step commitment inside the first call, and read both as diagnostics on whether a purchase exists rather than as steps in a persuasion sequence. Ask what has to happen internally for a purchase to be approved, and treat a vague answer as a finding rather than as something to be smoothed over. Verify consensus at the group level, not by summing individual enthusiasm, and expect that roughly three in four buying groups are carrying real internal conflict you have not seen. Attach a source, a date, and a falsifier to every qualification claim, and re-verify anything older than a few weeks in a buying group that large.

And render three verdicts, not one. Whether a purchase will happen. Whether it will be yours. Whether it is worth the hours. The frameworks in widest use are good at the second question. The data says the first one is where your deals are dying.