Activity is the easiest thing in a pursuit to mistake for progress, because activity is the thing that gets recorded.
A deal with eleven touches in three weeks looks alive in every system that watches it. The calendar is full. The thread is long. The stage advanced. None of that distinguishes between a customer pulling a decision toward themselves and a seller pushing one uphill, and those two deals have almost nothing in common except their appearance.
The pull audit is a way of separating them. It asks one question about every recent event in the deal: who initiated it?
Push and Pull Produce Identical Records
This is why the distinction survives so long undetected.
If the seller proposes a technical session and the customer accepts, the record shows a technical session. If the customer's architect asks for one because their team hit a question they cannot answer, the record also shows a technical session. Same artifact, same stage movement, same entry in the activity feed. The second deal is dramatically healthier than the first, and nothing in the shape of the data says so.
Run the audit and the difference becomes obvious immediately, because it is not subtle once you are looking for it. Take the last five meaningful events in the deal. For each one, write down who moved first. Not who benefited, not who ran the meeting: who reached out, who asked, who set the date.
Five out of five seller-initiated is a push. It does not mean the deal is lost. It means the deal is currently being carried, and it will stop moving the moment the seller stops carrying it.
Pull Is Costly, Which Is What Makes It Informative
The reason customer initiation is worth so much as a signal is that it costs the customer something.
Sending a calendar invite inside a large company means spending attention that was allocated elsewhere. Bringing a colleague into a conversation means putting a small amount of personal credibility behind the idea that this is worth their time. Asking for a document means committing to read it. These are not enormous costs, but they are real, and people do not pay them for things they do not intend to pursue.
Enthusiasm costs nothing. A buyer can say the demo was excellent, that this is exactly what they have been looking for, that they will take it to their leadership, and pay nothing at all for saying it. This is the mechanism behind the deal that stays warm for two quarters and dies without a decision. It was never lukewarm. It was cheap.
The audit weights signals by what they cost the person sending them.
The Audit Changes the Move, Not Just the Forecast
Discovering a deal is all push is only useful if it changes what happens next, and the instinctive response is the wrong one.
The instinct is to push harder: more follow-up, a better deck, an executive on the next call. That treats low pull as an effort problem. It usually is not. Low pull almost always means the deal is being worked with someone for whom the problem is not urgent, or not theirs.
The corrective move is lateral rather than harder. Find the person for whom this problem is already costing something and see whether they will spend anything on it. That test is small and it is fast: ask for something that requires the other side to move first. A question you need their answer to. An introduction only they can make. A date only they can set.
If they pay, there is pull. If they do not, you have learned that in a week rather than a quarter, and the ask itself did no damage.
In the book's fictional accounts, the pull audit does most of its work in exactly this way: not by predicting the loss, but by making the seller ask for something small enough to be answered quickly and consequential enough that the answer means something.
Two Ways to Read the Result Wrong
The first is treating one pulled event as pull. A single inbound question from a champion who has been quiet for a month is not momentum; it is one data point, and it may be a favor. Pull is a pattern across events and, ideally, across people.
The second is reading the audit as a verdict. A push deal early in its life is normal: nobody pulls toward something they have not yet understood. The audit is not asking whether pull exists today. It is asking whether pull is increasing. A deal in month three with the same push ratio as month one has not been progressing. It has been maintained.
The Question Underneath
Every qualification framework is ultimately trying to answer whether the customer will act without you in the room.
The pull audit answers it directly, using evidence the deal has already generated, requiring no new discovery call and no additional access. It costs one honest pass through the last five events and the discipline not to flatter the result.
Most deals will not survive that pass cleanly. That is the point. The ones that do are the ones worth the rest of the quarter.