Companies usually explain their wins by their strengths and their losses by price. Buyers often tell a different story. Win/loss analysis asks the people who made the decision why they chose you or someone else, and it frequently reveals reasons that never appear in CRM loss fields: a confusing proposal, a competitor's clearer message, a missing integration, a slow response at a critical moment. It does not need a large research budget. About ten well-run interviews are often enough to see the main patterns. This article explains how to select deals, who should interview, what to ask, how to analyse the answers and how to act on them.
Why ten interviews can be enough to start
Win/loss analysis is qualitative: the aim is to understand reasons, not to produce statistically precise percentages. In a set of interviews with buyers in your target market, the main themes often start repeating after several conversations. Ten interviews, split between wins and losses, are usually enough to identify the strongest patterns and decide what to investigate further. If the patterns are unclear or differ widely by segment, add more interviews in the segments that matter most. Treat the first round as the start of a continuous programme rather than a one-off study.
Selecting deals
Choose recent decisions, ideally closed in the last three months, while buyers still remember the details. Pick deals from your target segments, of meaningful size, where you reached a late stage such as proposal; early-stage losses tell you more about qualification than about competition. Include a mix of competitors and reasons recorded in the CRM, and if you lose often to doing nothing, include some of those. Avoid choosing only friendly customers or only the most painful losses. A balanced set of five wins and five losses gives you a comparison.
Who should interview
Buyers are more candid with someone who was not part of the sales process. A neutral interviewer, such as someone from marketing, product, a leadership team member not involved in the deal or an external researcher, usually hears more than the account executive would. Ask for 20 to 30 minutes by video or phone, explain that the purpose is to learn rather than to reopen the deal, and offer a small thank-you where appropriate and permitted. Record with permission, so the interviewer can listen rather than take notes.
The interview script
Use open questions and follow-ups. Let the buyer tell the story in order.
| Topic | Questions |
|---|---|
| Trigger | What was happening that made you start looking? Why then? |
| Problem | How would you describe the problem you were trying to solve? Who else cared about it? |
| Options | Which options did you consider, including doing nothing or building it yourselves? How did you find them? |
| Criteria | What mattered most in the decision? Did that change during the process? |
| Process | Who was involved, and how was the final decision made? |
| Experience | How did each vendor's sales process feel? What stood out, good or bad? |
| Decision | What was the deciding factor? What nearly changed your mind? |
| Advice | If you were advising us, what would you tell us to change? |
Coding answers into themes
After the interviews, read the transcripts and tag each reason, objection and decision factor with a short code, such as integration gap, implementation concern, clearer competitor message, price relative to value, responsiveness, trust in proof, or internal champion. Count how often each theme appears in wins and in losses. The most useful insights are themes that appear mainly in losses, or mainly in wins, because they distinguish outcomes. AI tools can help tag transcripts quickly, as long as you check a sample; the article on AI for customer research covers the method.
Reading the results honestly
Be careful with price. Buyers often cite price because it is easy to say, when the real issue is that the value was not clear enough to justify it. Look for what sits behind price: did the buyer understand the outcome, trust the proof, or see a cheaper option as good enough? Also look at the sales experience: slow responses, generic demos and unclear proposals frequently appear in losses. And note what winners valued most, because that is your strongest positioning material, often described in words you can use directly.
Turning findings into action
Share a short summary with sales, marketing, product and leadership: the top themes in wins and losses with quotes, and three to five recommended changes. Typical actions include rewriting positioning around the problem buyers describe, adding proof for the concern that most often blocks deals, changing a step in the sales process, adjusting packaging, or building a missing integration. Assign owners, and repeat interviews each quarter to see whether the themes change. Add a few structured questions to your CRM's closed-lost process, informed by the themes, so future data is more useful.
Common mistakes
Letting the rep who ran the deal do the interview, which tends to produce polite rather than honest answers. Leading questions such as was it mainly price. Interviewing only losses, which shows what went wrong but not what distinguishes wins. Waiting too long after the decision. Treating the analysis as a one-off. And collecting insights that never reach the people who could act on them. A simple, regular programme avoids most of these.
Getting buyers to agree to an interview
Response rates improve with a clear, low-pressure request. Ask soon after the decision, by email from someone senior or neutral, explaining that you are trying to improve how you work and want twenty minutes of honest feedback, with no sales follow-up. For lost deals, acknowledge the decision and make clear you are not trying to reopen it. Offer a few time options or a booking link. Where your policies and the buyer's allow, a modest thank-you such as a charity donation can help. If the decision maker declines, ask whether another person involved in the evaluation would speak with you; their view is often just as useful.
Making win/loss a continuous programme
After the first round, set a cadence that fits your deal volume: for example, two or three interviews a month, selected from recent closed deals across segments. Keep a single document of themes, updated after each interview, so that changes over time are visible. Review it quarterly with sales and marketing leadership and record the decisions it informed. Over a year, a continuous programme builds a much richer picture than an annual study, and it lets you see whether changes to positioning, pricing or the sales process actually shift the reasons buyers give.
Interviewing churned customers too
Win/loss analysis focuses on purchase decisions, but customers who leave after buying offer a related view. Interviewing a few churned customers shows whether the reasons buyers chose you held up in practice, and whether the problems that drove them away could have been seen during the sale, such as a poor fit that sales overlooked or expectations set too high. Use a similar script: why they bought, what they expected, what happened, when they started to doubt the decision, what they use now and what would have kept them. Combine these findings with win/loss themes, because together they show both how to win the right customers and which customers not to pursue.
FAQ
About ten, split between wins and losses, is often enough to see the main patterns. Add more where themes are unclear or differ by segment.
Someone not involved in the deal, such as marketing, product, a leader outside the sale or an external researcher. Buyers tend to be more candid.
Within about three months, while the buyer still remembers the details of the process and decision.
Probe it. Price is often cited when the underlying issue is unclear value, weak proof or a cheaper option seen as good enough.
Share themes with quotes across teams, agree three to five changes with owners, update CRM loss fields and repeat interviews quarterly.
It helps. Churn interviews show whether the reasons customers bought held up, and whether poor fit could have been spotted during the sale.
Yes, by tagging themes and pulling quotes from transcripts, as long as a person checks a sample against the originals.
Ask another person involved in the evaluation, keep the request short and neutral, and make clear you will not try to reopen the deal.