You Lost the Deal. Nobody Asked Why.

The deal goes quiet.

Not rejected. Not lost to anyone you can name. It just stops. The replies get shorter, then they get slower, then there is a polite note about circling back next quarter, and then there is nothing at all. Three months later somebody notices it still sitting in the pipeline and closes it out.

So you get the team together and you run a post-mortem, which in most small firms means everyone offers a theory. The salesperson thinks they went cheaper. You think the proposal was too long. Somebody who was never on a call is fairly sure they were only ever gathering quotes. Nobody knows. Then one person opens the CRM, picks a reason from a dropdown, types the word "price," and that is the last time anyone in your company thinks about that deal.

It gets worse. When Matthew Dixon and Ted McKenna ran machine analysis across 2.5 million recorded sales conversations for The JOLT Effect, they found that somewhere between 40 and 60 percent of qualified deals are lost to no decision at all. Not to a competitor. Not to a considered no. The buyer simply never moves. LinkedIn put the same figure at 40 percent at its Indie Summit this year, and pinned the cause on a buying group where nobody actively objects and nobody is convinced enough to push.

So for something like half the opportunities you work, there is not even a rival to blame. There is nothing obvious to investigate, so nobody investigates it.

You are running a business where the single most expensive event in the sales cycle happens regularly, and the only record of why is a guess.

"Price" Is Not a Reason. It Is a Polite Exit.

Here is what makes the dropdown so useless. Suppose you do pull the report and it tells you half your losses were marked "price." What is the action? Cut your rates by half? Send a coupon? You cannot do anything with it, because "price" is not a finding, it is a category.

Ask a buyer what "too expensive" actually meant and the answer is almost never about the number. It is closer to this: the money was findable, and nobody gave me a reason to go and find it. That is not a pricing problem. That is a value problem wearing a pricing costume, and the two get fixed in completely different departments.

The reason buyers reach for it is that it is frictionless. Saying "your onboarding sounded like a nightmare and your reference customer was nothing like us" invites an argument. Saying "it wasn't in the budget this year" ends the call in ten seconds. Your rep writes it down, and a fiction enters your CRM as a fact.

Scott Gillum spent seven years studying more than 10,000 buyers across 15 industries for The Hidden Buyer Journey, and found something that should change how you read every loss report you have ever seen. The reasons buyers give for choosing a vendor, things like product fit and reliability, are things they can only assess after the decision. They are not the drivers. They are the justification assembled afterwards. What actually moves the decision is whether the buyer felt confident, whether the company seemed credible, whether the risk of being wrong felt survivable.

Which means even an honest buyer, trying hard to help you, will hand you a rationalization. Getting past that is a skill, and it starts with asking better questions than the dropdown asks.

Four of them turn a vague reason into something you can act on:

• Too expensive, or never worth it? One of those is a number. The other is that nobody built the case.
• Missing a feature, or missing an explanation? Firms lose deals every week over something they actually do, and never said clearly enough for anyone to notice.
• Beaten by a better competitor, or a faster one? Turning up first with a simple story beats turning up third with a better product more often than anyone wants to admit.
• Not ready, or never given a reason to move? "Let's revisit next quarter" is what a buyer says when nothing about this quarter felt urgent.

Each pair sounds like one question and hides two. One half costs you a product roadmap or a discount. The other half costs you an afternoon and a rewritten page. We have written before about how a good meeting dies in the rooms you were never invited to, and this is the same failure seen from the other end: you were not there, so you are guessing.

The One Person Who Cannot Get the Truth Is the One Who Was There

The obvious move is to have the salesperson ask. It is the obvious move and it does not work.

Think about it from the buyer's side. They told your rep no. A week later that same rep wants half an hour to discuss the decision. There is exactly one thing the buyer assumes is happening, and it is that they are about to get sold to again. So they either decline, or they turn up and give you the smoothest, least useful version of events. It was a budget thing. Timing was off. Great process, honestly.

People who do this work for a living describe it as the difference between the breakup conversation and the one she has with her best friend afterwards. To your face it is not you, it is me. To someone neutral, the actual list.

Every serious source on this says the same thing: the person who ran the deal should not be the person who reviews it. Then every one of them assumes you have a product marketing team, a competitive intelligence function, or the budget for an agency. Outsourced win/loss programs commonly run twenty to forty thousand dollars for a three-month pilot, and ninety to a hundred and twenty thousand a year once a company commits to an ongoing one. That is a real service and it earns its money at a certain size. It is also completely irrelevant to a firm with nine people in it.

So here is the translation nobody bothers to write down for small firms. You do not need a neutral department. You need a neutral posture, and an owner can hold one more easily than a commissioned salesperson can. When the founder calls and opens with "I am not here to sell you anything, you already made your decision, I am trying to understand it so we get better," that is credible in a way it never is coming from the rep whose number depended on the outcome. Founders get a hearing that staff do not. It is one of the genuine advantages of being small, and almost nobody uses it.

You Are Not Investigating Sales. You Are Doing Market Research.

The reason this gets shelved is that it looks like a sales exercise, so it gets filed under sales, where it competes with hitting this quarter's number and loses every time.

It is not a sales exercise. Look at what actually comes out of it. You learn which kind of company says yes quickly and which kind drags you through four months and disappears, which tells you who to target. You learn the words buyers use for the problem, which is what should be on your homepage instead of the words you use for your solution. You learn which of your competitors keeps turning up and what they say about you when you are not in the room. You get the one thing no attribution report can ever give you: the buyer telling you, in their own words, what actually moved them. You learn whether the case study you keep sending is landing or being ignored. Every one of those is a marketing decision, made with evidence, in a business that otherwise makes them on instinct.

And there is a second payoff that gets missed because it is not dramatic. Everybody wants to interview the losses, because losing stings and stings demand explanation. The wins are the underrated half. They are far easier to get on a call, they are in a good mood, and they will tell you the thing you are doing right that you did not know you were doing. Firms go looking for a weakness to fix and find a strength they had no idea was carrying them, one they had never once mentioned on their website.

The finding that comes up most often across this kind of work is not "your product needs another feature." It is "some of your customers love you and some of them complain constantly, and they are two different kinds of company." The fix is not a roadmap. It is narrowing who you go after. That is a positioning decision, and it is the same argument we made about why most B2B positioning fails: the problem is rarely the offer, it is the aim.

Run It Backwards, and Run It Narrow

Most attempts at this die because somebody tries to build a program. An ongoing monthly process, a dashboard, a workflow, sign-off from three people. It never launches.

Do the opposite. Run a single sprint, six weeks, and point it at the past rather than the future. You already have two quarters of closed deals sitting in your CRM, won and lost, and not one of them has been touched. You do not have to wait for new deals to close to start learning. Going backwards gives you a bigger pool to draw from, produces an answer inside six weeks instead of six months, and if it turns out to be useful you have proven that before asking anyone to commit to it permanently.

Then narrow it. The instinct is to ask "why do we win and lose," and that question is unanswerable. You will pull deals of every size across every industry, get a shapeless pile of anecdotes, and be unable to act on any of it. Pick one slice instead. Mid-sized manufacturers. Or every deal over 25,000 dollars. Or every deal where the same competitor showed up. One slice, one question, one output.

Aim for a rough balance of wins and losses. Losses are harder to book, which tempts people to give up on them and end up with a file full of happy customers. Chase the losses harder, not less.

And move quickly. The window is about four weeks from the deal closing or collapsing. Interview a win after they have been using you for two months and they will describe why they chose you through the lens of how it has gone since, which is a different question. Interview a loss after they are six weeks into a competitor's product and their memory of your process is already colored by whether that product is working out.

Getting Somebody to Actually Say Yes

This is the step that quietly kills most attempts, because people send one email, get no reply, and conclude that buyers will not talk to them.

Buyers will talk to you. Recruiting for this is a sales motion and it has to be worked like one. The numbers to plan around: expect roughly one in five to agree, so if you want five conversations, invite twenty. And more than half the acceptances arrive after the first email, which means a single send is not a test of anything.

A sequence that works:

• The first ask comes from whoever they actually know. The person they emailed with for six weeks, not a name they have never seen.
• A short follow-up three or four days later. Nothing clever. Just that you would still value twenty minutes.
• Offer something for their time. A gift card is normal and it is not a bribe, it is an acknowledgment that their time has value.
• The last email is the one that works. "Should we mark you as not interested?" It gives them a clean way out, and a surprising number of people who never responded will suddenly respond to it.

Say twenty minutes and mean it. Ask for the person who actually decided, not the person who was easiest to reach. Ask permission to record, because you cannot listen properly while taking notes, and listening is the entire job.

One more thing, and it matters more than the rest of the sequence. When you invite a lost deal, say in the email that this is not an attempt to win the business back. Any suspicion of that and they will not come, and if they do come they will not be honest.

The Questions, and the Rule That Beats All of Them

Walk them through their own buying process in order, and the answers arrange themselves.

• What was going on that made you start looking at all? Why then and not a year earlier?
• Once you decided to look, how did you go about it? Who else did you talk to?
• What were you comparing on? What actually mattered by the end?
• Who else was involved in the decision internally?
• What tipped it? If you can name the two or three things that decided it, what were they?
• Set the product aside. How was the experience of buying from us?
• What did you need at some point in that process that you could not get from us?

Then the closing question, which routinely produces more than the rest of the list combined: if you were running my company, what would you change to serve people like you better? It moves the buyer out of the seat of the person being interviewed and into the seat of someone giving advice, and people are far more candid, and far more specific, when they are giving advice.

Now the rule that matters more than any of it: this list is a guide, not a script. The value is not in getting through seven questions. It is in the moment somebody hesitates, or says "well, sort of," or laughs before answering. That is the thread. Pull it, and let the rest of the list go. A conversation that never got past question two, because question two turned out to be where the whole thing went wrong, is a successful interview.

You Are Done When the Answers Stop Surprising You

The first objection inside the company will be that this is not a big enough sample to mean anything. Somebody will say the words "statistically significant," and it will stall there.

That standard belongs to survey data and this is not survey data. The standard for conversations like these is saturation: you keep going until new interviews stop producing new themes. When the fourth person in a row describes the same hesitation you heard from the first three, you do not need a fifth to confirm it, you need to go fix it. In one narrow slice, that usually lands somewhere between ten and fifteen conversations, and sometimes fewer.

The narrower your slice, the faster you get there. This is the practical reason to resist the urge to interview everybody.

One Page, or It Did Not Happen

The failure at the end is a folder of recordings nobody opens. Everything you learned is technically saved and functionally gone.

The output is one page. Six things on it:

• Scope. Which deals, which segment, what period.
• The pattern. The one thing you keep hearing.
• The evidence. Two or three quotes, in the buyer's words, not paraphrased.
• The likely cause. Your read on why it is happening.
• The one action. What changes because of this. Not a list of nine. One.
• The owner and the date. Who is doing it and when you look again.

Put that page in front of people at a moment when decisions are actually being made, not at a random Tuesday meeting. And keep the old ones, because the most useful signal arrives later: if the same theme comes back next quarter, word for word, you are not learning anything new. You are collecting findings and failing to act on them, which is a more expensive problem than not having asked in the first place.

The Tape Does Not Care How You Felt About the Game

Every professional team, in every sport, watches the tape back. Not to relitigate the loss. To find the pattern nobody could see from inside the play.

Selling is the same, and small firms almost never do it, because the deal is over and there are new ones to chase and asking feels a little like reopening a wound. But the cost of not asking compounds quietly. You lose the same way to the same objection for two years, spend money making the wrong argument to the wrong companies, and never learn that the thing your best customers value most is not even mentioned on your website.

Ten conversations. A few gift cards. One page at the end. There is no cheaper research available to a small business, and there is nothing else that tells you the truth about how you are actually perceived by the people whose opinion decides whether you grow.

If you want a second pair of eyes on what those conversations turn up, or help turning them into positioning and campaigns that reflect what you learn, that is the sort of thing we do. But run the interviews either way. The answers are sitting in a dozen inboxes right now, and every one of those people would probably tell you, if somebody asked.