Win/loss analysis: How to run one that changes what you do next

Win/loss analysis works when it runs every month on every closed deal above a threshold, with the calls made within 4 weeks of the decision by somebody who was not on the deal. Run that way it produces a dataset you can track, rather than a document nobody opens twice.

Almost every company at 40 to 200 people has an opinion about why deals are won and lost. Almost none has evidence. The opinion comes from the sales team, which is the one group in the building with the least ability to hear the real answer.

Key takeaways

• Monthly cadence beats an annual study, because the point is spotting a change early.
• Call within 4 weeks. After that the buyer has rewritten the story in their own head.
• The person who worked the deal should never make the call.
• Price is the most reported loss reason and rarely the real one.
• Code the answers consistently, or you have anecdotes instead of a trend.

Run win/loss every month, on every closed deal above a threshold

Set the scope once and stop debating it. Cover both outcomes, wins as well as losses, because wins tell you what to repeat and losses tell you what to fix, and having only one gives you a skewed picture.

Set a value threshold at a deal size worth 30 minutes of somebody's time, and rely on the sales notes below it. Review every qualifying deal rather than a sample, since choosing which ones to look at reintroduces the bias you are trying to remove.

Then hold a monthly rhythm. 4 to 6 calls a month is enough at this company size, and it means you notice a change in the market inside a quarter.

This is different from the one-off interview round that sets your positioning. That exercise designs a sample across wins, expansions, losses, and churn to answer who the best customer is. Win/loss analysis runs continuously against the sales process, and the two feed each other.

Call within 4 weeks, and never let the rep who lost make the call

Two mechanics decide whether the answers are any good.

Timing. Aim for 2 to 4 weeks after the decision. Earlier and the buyer is still in procurement mode. Later and they have tidied the reasoning into something simpler than what actually happened, usually price.

Who calls. Somebody other than the person who worked the deal. Buyers soften the answer to avoid an awkward conversation, and the rep hears confirmation of what they already believe. Use someone senior from outside the deal, or a partner. The opening line that works:

• Say you are not trying to reopen the decision.
• Say the call is 20 minutes and the notes stay internal.
• Ask for help improving, which most buyers are willing to give.

Expect roughly half of lost prospects to agree at this deal size. That is enough. A refusal is data too, especially when it clusters around one competitor.

Why "we lost on price" is almost never the real reason

Price is the easiest answer for a buyer to give and the easiest for a rep to accept. When you ask properly, the actual reasons usually turn out to be one of these:

• Perceived risk. Nobody could tell whether the switch would break something. A missing case study closes that gap faster than a discount does.
• No internal champion. Your contact liked it and could not carry it to a committee. They needed a one-page business case you never gave them.
• Timing. Budget cycle, a reorganisation, a bigger project ahead of it. Worth a callback date rather than a loss.
• Missing proof for their situation. You had case studies from a different industry or company size.
• Genuine price. It happens, and it is usually a smaller share than the CRM suggests.

The test: ask what would have had to be true for you to choose us. Buyers answer that question honestly, and the answer is rarely a lower number.

Code every answer the same way so trends become visible

The output is a dataset, so it has to be recorded consistently.

• One primary reason per deal. Forced choice from a fixed list of 6 to 8, with a free-text note for detail.
• Record the alternative they chose. Including internal builds and doing nothing, which are competitors most companies never track.
• Capture the exact wording. How the buyer described the problem, which is the part your website can use.
• Keep it in the CRM against the deal record, so it survives the person who collected it. The GTM modernization after an acquisition sequence covers getting the record into that shape.
• Review the totals quarterly. Monthly numbers at this volume are noise. A quarter of coded calls shows a pattern.

Who acts on each finding, and what they change

Findings with no owner turn into a report, so route each reason to the team that can fix it.

Perceived risk belongs with marketing, which answers it with case studies from the same industry and size, plus an implementation page that addresses what could go wrong. A missing champion belongs with sales enablement, as a one-page internal business case the buyer can forward to their own committee.

The wrong buyer belongs with positioning. If you keep losing a segment you keep chasing, the target needs revisiting, and the portfolio company positioning play covers how that gets fixed. Feature gaps belong with product, with the deal value attached so the roadmap conversation has a number in it. And timing losses go back to sales as a dated callback rather than a closed file.

If you only have 1 or 2 marketers: 5 calls a month, 3 hours

You do not need a programme. You need a habit that survives a busy month: 5 calls a month across a mix of wins and losses, made by one senior person, 20 minutes each.

Write one page per call with the coded reason, the alternative they chose, and 2 verbatim quotes. Then write one quarterly summary with the coded totals and 3 recommended changes.

That is about 3 hours a month, and it is the cheapest research any company at this size can run.

What the calls revealed in a 60-person Nordic software company

Take a newly acquired Nordic software company we work with: around 60 people, 600+ customers, two marketers, no CMO. The CRM recorded most losses as price.

The calls said something different. The larger buyers the company was now chasing worried about disruption during the switch, and nothing on the website addressed it. The most common alternative was an internal build, which appeared nowhere in the competitor tracking. Several buyers liked the product and could not get it through their own committee without a business case. And genuine price losses were a minority, mostly in a segment the company was leaving behind anyway.

The fix list came back as 3 marketing items rather than a discount policy.

FAQ

Start with 5 calls this month

Pick the 5 largest deals that closed in the last month, wins and losses, and get someone senior on 20 minutes with each. Code one reason per deal, keep the verbatim wording, and route each finding to whoever can act on it. Then do it again next month. No random acts of marketing.

If the pattern that comes back suggests the message is aimed at the wrong buyer, that sits under the value-creation programme for portfolio companies.

When you are ready to set this up for a company you have just bought, book a strategy session. You will leave with a bird's-eye plan, whether or not we work together.

Book a strategy session →