# Win-loss analysis for a small team: ten deals, five questions

https://salesbond.app/blog/win-loss-analysis-small-team-pipedrive/
Published: 2026-10-01 · Updated: 2026-10-25

Win-loss analysis means looking at the deals you won and lost through the buyer’s eyes, not through the reason code your CRM stored. For a small team, that comes down to ten recent deals and five questions. Pull the last ten closed-lost deals out of Pipedrive, call each buyer and ask why they chose what they chose, then put their answer next to the reason your rep logged when they marked the deal lost. The gaps between the two columns are the findings. No research vendor, no quarterly program: one afternoon of calls and a single page at the end.

## Why loss reasons mislead

Filter Insights by lost reason and the same two bars tend to come back quarter after quarter: no budget and bad timing. That is not a market insight. The lost reason in your CRM is the rep’s self-report, typed in the thirty seconds after a deal died, and [Sendspark’s guide to win-loss for B2B sales](https://blog.sendspark.com/win-loss-analysis-b2b-sales) makes the same observation: reason codes skew toward exactly those two answers. Both are true often enough to be plausible and vague enough that nobody can argue. Neither tells you what to change on Monday.

Part of it is how the field works. By default it’s a freeform box the rep fills in at the moment of marking the deal lost, and a rep who just lost has little incentive to write “I never got to the CFO” when “budget” is right there. A predefined list makes the answers tidier ( [we covered that setup in Pipedrive lost reasons](https://salesbond.app/blog/pipedrive-lost-reasons/)), but it still records the rep’s version. The buyer’s version lives in the buyer’s head, and the only way to get it is to ask.

So treat the CRM field as the hypothesis column. The buyer call is the test.

## Pick the ten deals

Most closed-lost analysis stops at the CRM export. This one starts there.

Take the last ten deals marked lost where the buyer actually chose something: a competitor, an in-house workaround, or keeping the old process. Skip deals that went dark before a demo and deals you disqualified yourselves. Those belong in a different conversation (we cover it in [when to close a deal as lost](https://salesbond.app/blog/when-to-close-a-deal-lost/)). You want cases where a real decision was made by a person who can still remember making it.

In Pipedrive this is a filter on the deals list: status lost, lost date within roughly the last ninety days, sorted by close date. Add the lost reason column to the list view so you can read the rep’s reason for all ten at once before a single call. Copy that view into a sheet with four columns: deal, rep’s reason, buyer’s reason, gap. If you prefer to read them on the board, Pipedrive can now also keep [closed deals in the pipeline view](https://salesbond.app/blog/pipedrive-closed-deals-toggle-weekly-review/) with a toggle.

Why ten and not thirty? A team of five reps will not finish thirty interviews, and a half-finished sample is worse than a small finished one. Ten calls is one afternoon for one person. You are comparing, not doing statistics, so look for something that shows up three or four times rather than for a percentage.

Before you call, do what [ZoomInfo’s Pipeline blog](https://pipeline.zoominfo.com/sales/wringing-the-most-out-of-winloss-analyses) recommends for any win-loss review and make the records comparable: the same stage history, reason code, notes, contacts and timeline for every deal. In practice that means ten minutes per deal with the notes and activities, writing down who the rep talked to, which stage the deal stalled in, and how long it sat there. You need this on the call. The buyer will say something like “we never saw pricing,” and you want to know whether the activity log agrees.

## Five win-loss interview questions to ask the buyer

Who calls matters. If the rep who lost the deal calls, the buyer will be polite and the answer will be “budget” again. ZoomInfo’s guide assumes a neutral interviewer. A small team does not have one, so the closest substitute is the sales lead or the founder, introduced honestly: “You chose someone else, we are fine with that, and I would like fifteen minutes to learn what we got wrong.” In our experience buyers who went elsewhere are glad to talk. They were rarely asked.

Then ask these five, in this order, and resist the urge to add a sixth.

1. What did you end up doing, and what made that the right call for you?
2. At what point in our process did you stop seriously considering us, and what happened around then?
3. Who else in your company was part of the decision, and what did they care about that we never addressed?
4. What did the option you chose do or say that we did not?
5. If we had done one thing differently, what would it have been?

Question one is the buyer’s version of the lost reason, in their words. Question two gives you a stage, which you will compare to the stage history in Pipedrive. Question three is where “no budget” usually turns into “the finance lead never saw a business case.” Question four is the only competitive question, and it is deliberately about behavior rather than features. Question five gets the most candid answers, because it sounds like a request for advice. And when question one turns up [a missing feature the buyer named](https://salesbond.app/blog/feature-requests-pipedrive-deals/), log it on the deal too.

Write the answers down close to verbatim. Paraphrasing is how “your onboarding looked like a six-week project” becomes “implementation concerns” and loses everything useful.

The answers about why the buyer picked a rival are the raw material for a [sales battlecard](https://salesbond.app/blog/sales-battlecards-small-teams-pipedrive/), so keep them in one place instead of in call notes.

Hold the call to fifteen minutes. Don’t pitch, don’t correct the buyer’s memory, don’t ask whether they would reconsider.

## Reading rep reason vs buyer reason

Now the sheet has ten rows and the real work is one hour at a table with the whole team.

![Ten-row win-loss sheet comparing the rep's lost reason with the buyer's reason, three mismatched rows highlighted](https://salesbond.app/wp-content/uploads/2026/10/win-loss-analysis-small-team-pipedrive-2.png)

Look for three things, in this order.

Direct contradictions first. Rep wrote “price”; buyer said “we never understood what we were paying for.” Rep wrote “timing”; buyer said “your rep stopped following up after the second call.” These are the loudest findings and usually the most uncomfortable, which is why they belong in a room rather than in an email thread.

Then the stage. Answers to question two cluster. If most buyers say they checked out between the demo and the proposal, you have a proposal problem no matter what the CRM field says. Match the buyer’s stage to the stage history in Pipedrive. A deal that sat in “Proposal sent” for three weeks with no activity logged gives you the mechanism as well as the symptom.

Last, the missing person. Count how often question three names someone who never appears among the deal’s contacts in Pipedrive. That count is your single-threading problem in plain sight, and it tells you more than anything the reps typed into the field.

Then write one page. Not a deck. The page has the pattern in one or two sentences, the deals that show it by name, the quotes that make it hard to dismiss, and the one change you are going to try. Below that goes a short list of things that came up once and are probably noise. A small team will read a page at the next pipeline meeting. It will not read a program.

## What to change and when to repeat

One change per round, taken from the losses; the wins get their own short round later. The temptation is to fix everything the buyers mentioned, and then nothing changes. Pick the pattern that appeared most often, decide the the smallest change that addresses it, and put it where reps will meet it: a required field at the stage where deals died, a line in the [deal review questions](https://salesbond.app/blog/deal-review-questions-buyer-action/) you already ask, a proposal template that opens with the business case for the person who was never in the room. If the pattern points at the first conversation, the change can be to [test a new sales pitch version](https://salesbond.app/blog/sales-pitch-small-team-pipedrive/).

Then let the CRM field catch up with what you learned. If buyers told you the real reasons were “no business case for finance” and “stayed with the incumbent after the trial,” add those to the predefined list so next quarter’s lost reasons have a chance of being true. Keep the list short, a long list gets the same lazy pick as a freeform box.

Repeat when you have ten new losses, not on a calendar date. For a team closing a handful of deals a month that is about a quarter; with lower volume, longer. In the second round the question is whether the top pattern shrank. That is also the moment to look at [win rate by stage](https://salesbond.app/blog/sales-win-rate-pipedrive/), because the first thing a good change does is move where deals die, not how many. (The two numbers get mixed up: win rate is deals won divided by all closed deals, while a win-loss ratio is won divided by lost.)

The wins come after the second loss round. Five calls, the same questions reworded for someone who said yes. Winners tell you what to keep doing, and a small team forgets that faster than it forgets the losses.

If ten calls feels too small to call analysis, start anyway. This week, save the lost-deals filter in Pipedrive, add the lost reason column, and book the first buyer call for Friday.