Picture the call you are asking for.
Your salesperson rings the buyer who turned them down last week and asks, politely, what went wrong. What they are actually asking, and both people on the call know it, is whether it was their fault.
So the buyer gives them the kind answer. Not a lie, just the one that ends the conversation without anybody feeling bad, and that answer is almost always price.
Then the reason goes into the CRM as “price”, and six months later somebody reads a report saying you lose on price, and the response is to become cheaper. A flaw in who did the asking has turned into a margin decision.
The method predetermined the finding, and the finding cost you money.
So the fix is not more discipline about post-mortems. It is separating the person who asks from the person who owned the deal, and then asking the won deals too, which almost nobody does and which is where the useful material actually is.
The one number worth knowing
There is a figure that makes this concrete, and a caveat that has to come with it.
User Intuition, analyzing their own lost-deal interviews, found price was the primary driver in 18.1% of losses while 62.3% of buyers initially cited price. The first answer and the real answer differ by a factor of three.
The caveat: every firm publishing numbers in this area sells win/loss research, so all of it points in a direction convenient to the seller. Clozd reports that buyer and seller explanations for a loss align only about 15% of the time, and that 44% of the closed-lost reasons sitting in CRMs are outcomes rather than reasons. Those are interesting and they are also marketing for a service.
I would not quote any of them as established fact. What makes the 62.3% against 18.1% gap worth repeating is that it matches what the structure of the conversation predicts: a buyer with no incentive to be unkind, talking to the person they rejected, reaches for the answer that is least personal. You would do the same.
On what “price” actually means when you hear it inside a live negotiation, I have written that up separately and will not repeat it here. This is the different problem of hearing it afterwards, from somebody being polite, and then believing it.
How much of this is even happening
Worth knowing how unusual it is to do this at all, though the figures disagree in an instructive way.
Clozd’s 2025 work puts the share of companies running an ongoing, cross-functional win/loss program at 39%, up from 30% the year before. Pragmatic Institute, back in 2023, put the share doing some form of win/loss analysis at 83%.
Both can be true, because they count different things. Most companies do something: a note in the CRM, a question on a pipeline call. Far fewer run anything that would survive being called research.
That gap is the whole subject. The informal version is not a smaller version of the real thing. It is the thing that produces the wrong answer, because it is conducted by interested parties with no protocol.
Who should ask
The requirement is simple and almost nobody meets it: the person asking must have nothing at stake in the answer.
In a company of any size, that is somebody from outside the deal. Marketing, operations, a founder who was not the seller, or an outsider. Not the account owner, and not their manager either, because a manager asking about a loss is conducting a performance review with a customer present.
Say so explicitly when you ask. “I’m not on the sales team and this isn’t about them, I’m trying to understand how you made the decision.” That sentence does more work than any question that follows it, because it removes the reason to be kind.
Then ask about the process rather than about yourselves. Who else was involved. What the alternatives were. What nearly changed the outcome. When the decision was effectively made, which is usually earlier than anybody on your side believes.
Avoid the question everybody opens with, which is “why didn’t you choose us”. It invites a verdict, and verdicts are where the courtesy answer lives.
Ask the won deals
Here is the part that gets forwarded, because it inverts the whole exercise.
Interview the customers who said yes. Almost nobody does, because a win does not feel like it needs explaining.
Three reasons it is better material. Winners will actually talk to you, at length, without discomfort, so your response rate is several times higher. They will tell you what nearly stopped them, which is the objection your next ten prospects are having silently. And they will tell you what they were comparing you to, which is frequently not who you think your competitors are.
You will also find out what they think they bought, which is often not what you think you sold. That one sentence has reshaped more positioning than any loss interview I have been involved in.
A practical ratio: three won deals for every lost one. The losses tell you where the leaks are, and the wins tell you what the product actually is.
The sample problem nobody mentions
One more trap, and it is the reason even well-run win/loss work misleads.
You can only interview the deals that got far enough to have a contact who will take your call. Those are the deals that reached a proposal, which means your research systematically excludes everybody who disqualified you earlier, and that group is larger.
The prospect who read your pricing page and left. The one who could not tell from your site whether you serve companies their size. They are not in your sample and they never will be, so win/loss analysis is structurally blind to the top of your funnel.
Which means you should not use it to answer questions about demand. It answers questions about competitive evaluation. Those are different problems, and running one piece of research for both is how a company concludes it has a sales problem when it has a positioning problem.
Where I come out
If you do one thing, do not start with the losses.
Call the last three customers who bought and ask what nearly stopped them, and who else was in the running. It takes an afternoon, the calls are pleasant, and you will learn more than a quarter of lost-deal post-mortems conducted by the people who lost them.
Then, for the losses, change who makes the call. Keep the question about process rather than fault, and treat the first mention of price as the beginning of the conversation rather than the end of it.
If you’d like help working out which of your own deals are worth a proper conversation, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on getting marketing and sales pointed the same way cover what to do with the answers. I’m taking on a small number of contracts at the moment, so I’ll tell you straight whether it’s something I could help with.