Win/loss analysis is one of the highest-ROI activities in B2B sales. It is also one of the most consistently mismanaged.

Gartner research summarised in Klue’s win-loss analysis statistics shows that companies with formal win/loss programs achieve 15 to 30 percent revenue increases, and up to 50 percent improvements in win rate. Organisations running the practice for two or more years report 10 to 20 percent sustained lifts in win rate. The financial case is not subtle.

And yet the typical win/loss program produces a quarterly PDF that nobody reads, written by a researcher in product marketing, based on twelve interviews none of which were with the deal team. The findings are interesting. The findings change nothing.

“Win/loss is not a research project that produces a report. It is a leadership instrument that produces decisions. If your program is producing the first, you do not have a program.” — Mark Southgate

The difference between the two versions is not the methodology. It is where the work sits in the operating cadence.

The fact that should bother every CRO

A piece of research from Corporate Visions, based on more than 100,000 B2B purchase decisions, found that sellers and buyers give different reasons for deal outcomes between 50 and 70 percent of the time. Half the time, the seller’s “we lost on price” was actually “we lost on the demo.” The seller’s “we won on relationship” was actually “we won because the incumbent fumbled the renewal.”

A more uncomfortable companion stat: 53 percent of buyers in a losing vendor’s deal say the vendor could have won if a fixable misstep during the sales process had been corrected. Roughly half the losses are recoverable on the dimensions sellers control.

If those two numbers are even directionally true in your organisation, the implications are large. The pipeline review, the deal review, and the forecast call are all running on seller interpretations of why deals are moving. Half the time, those interpretations are wrong. The leadership team is making decisions on data the buyer would dispute.

This is the gap a real win/loss program closes. Not by producing a report. By feeding corrected reality back into the operating cadence, deal by deal, week by week.

What a leadership-grade win/loss system looks like

Three components.

Component 1: structured buyer-side interviews on a meaningful sample. Not the deal team’s notes. Not a satisfaction survey. A 30 to 45-minute conversation with the actual buyer, ideally led by someone outside the sales team. The structure matters: a small number of specific questions, asked in the same order, comparable across interviews. Without that, you have a collection of anecdotes, not a dataset.

Component 2: a feedback loop into the deal review forum. This is the part most programs skip. The findings have to land in the place where deal strategy is set, not in a quarterly slide deck. The simplest mechanism is a standing 15-minute slot at the start of each deal review where the most recent win/loss findings relevant to the deal under discussion are surfaced. Five minutes per deal, once a fortnight, changes how the room reasons about its current pipeline.

Component 3: pattern tracking, not anecdote tracking. One interview teaches you about one deal. Twenty interviews teach you about your market — if the data is structured. Patterns to watch: which buyer roles consistently kill deals at which stage; which competitive losses recur; which product gaps appear in losses that the product team has not seen; which sales motions correlate with wins versus losses in a given segment.

The first time most companies do this rigorously, the findings are uncomfortable. Champions are weaker than the deal teams thought. Pricing pushback is more often about value framing than price level. The competitor everyone fears is winning less than the unknown competitor nobody is tracking. None of that lands as a report. It lands as a series of changes to how deals get reviewed and forecast.

“Half of what your sales team believes about why deals are won and lost is wrong. The half they have wrong is the half they would change if they could see it.” — Mark Southgate

What it costs to do nothing

The companies that skip a real win/loss program tend to share a pattern: forecast accuracy improves slowly, deal reviews repeat the same surface-level diagnoses, and the same competitive losses recur quarter after quarter without any structural response.

That last point is the most expensive. Win/loss done well surfaces patterns that no individual deal review can. A competitor that wins one in three deals against you in mid-market manufacturing is invisible in any single deal. It is loud in a pattern. The cost of not seeing the pattern is paying full price for losses you could have prevented by changing your competitive positioning, your sales motion, or your enablement.

As Mark Southgate notes: “The deals you lost last quarter are the cheapest market research you will ever have access to. Most companies throw the data away because they cannot stomach what it says.”

The simplest version that actually works

If your win/loss program does not exist yet, or exists only as a survey product marketing runs once a year, here is the minimum viable version that produces leadership-grade output:

  1. Interview the buyer on every deal over a defined threshold — typically every deal above your average enterprise ACV, plus a sample of mid-market wins and losses. Use an external interviewer; the data is materially better. Research from Klue shows companies using third-party interviewers are more than 2x as satisfied with feedback quality compared to internal programs.
  2. Use a five to seven question protocol. What problem were you trying to solve? Who were the alternatives? What made the difference? What would have changed your decision? Who was the most important voice in the room? What surprised you about our process? What would you tell us if you knew we would not get defensive?
  3. Surface findings in the deal review, not a separate forum. Bring the most recent five interview themes into the next deal review. Connect them to live deals.
  4. Review patterns quarterly with the leadership team. Not as a report. As a list of three to five things the company will change as a result.

The output of this version is not a deck. It is a small number of structural changes per quarter — to the sales process, the competitive positioning, the qualification criteria, or the deal review focus. That is what makes it a leadership system rather than a research artifact.

Tying it back to the operating cadence

Win/loss analysis only works when it lives inside the operating cadence. Sitting next to it, in a separate function, with a separate calendar, it becomes ornamental. Embedded inside the deal review and the quarterly leadership forum, it becomes one of the most powerful instruments the CRO has for shaping go-to-market strategy.

The 15 to 30 percent revenue lift the Gartner research describes does not come from running the interviews. It comes from acting on what they say, in the forums where action is possible. Most companies do the first half. The second half is what makes it a program.