What the founder never hands over

Before replacing the person, inspect the job

A VP Sales leaves after nine months.

The diagnosis starts before the door closes. Wrong profile. Too corporate. Too junior. A scaler when the company needed a builder.

Any of it can be true. I have seen each of those be true.

But all four explanations rest on the same premise, and nobody tests it: that there was a working sales motion, and this person failed to run it.

That premise is worth one morning of your time before the search reopens.

A CRM is not a sales motion

When a founder hands over sales, what moves is what was written down. Pipeline, accounts, contacts, the CRM, the deck, the price list.

Early deals are rarely won with only what was written down.

The founder knows which objection is real and which is theatre, when a roadmap promise is safe to make, how far a price can bend before anyone minds, which reference customer will actually pick up the phone. And underneath all of it: the founder can commit the company without asking permission.

That last one is not knowledge. It is authority, and it appears on no handover checklist.

So the new hire inherits the pipeline without inheriting the conditions that made the pipeline convertible. Then we measure them on conversion.

I learned this from the wrong side

At AiZimov I was the founder and the salesperson. I sold the first customers myself, then hired people to scale what I had been doing.

Handing them my accounts was not handing them my way of selling.

Some of what worked was knowledge I had never written down. Some of it was simply that I was the founder. I could change a commitment, take a commercial risk, decide in the moment. They could not, and nothing in their job description said so.

At the time I read that gap as an execution problem on their side. I was looking at the wrong thing.

AiZimov closed, for several reasons, funding among them. That does not prove founder dependency kills companies, and I am not going to pretend it does. What it taught me is narrower and more useful: a founder-led sales motion does not become transferable because somebody else gets the CRM login.

You can test this without another strategy meeting

Take the last ten deals the company won that mattered.

For each one, find the first moment the founder entered, and what the team needed from them when they did. Expertise. Credibility. A price outside the grid. A contractual decision. A product commitment. Access to a relationship.

Do not interpret yet. Put the ten side by side.

If they tell ten different stories, good. A founder showing up on important accounts is not a defect.

If the same intervention keeps appearing, you are not looking at a hiring problem. You are looking at a part of the sales motion that still belongs to one person, and no candidate fixes that by arriving.

Instead of asking what profile to hire next, ask what the last person repeatedly needed from the founder in order to close. That question does not absolve a weak hire, and it does not prove the founder is in the way.

It separates two problems that look identical from the outside and have nothing in common: someone who could not run a working sales motion, and a sales motion that was never made runnable by anyone else.

Hiring solves the first. It almost never solves the second.

What did the last person repeatedly need from you in order to close?

What I can’t do : I can't tell you from a table whether someone was a good VP Sales. Nobody can. What I can see is where closing repeatedly depends on one person, one relationship, or a decision nobody wrote down — and what would have to change for that to stop being true.

Isn't this just letting a weak hire off the hook?

No. The table cannot tell you whether someone was good. It can tell you whether the job had a repeating dependency in it. A weak hire and an untransferable job produce the same resignation letter, and they need opposite responses.

Our founder is deliberately involved in large accounts. Is that a problem?

Usually not. A founder on a strategic account is a feature. The question is whether the deal could have closed without them, and whether anyone has ever checked. Ten deals side by side answer that faster than a conversation does.

How is this different from a sales audit?

An audit produces a document and a proposal. This produces ten rows you fill in yourself, and either a repeating pattern or no pattern at all. If there is nothing in it, there is nothing, and I will say so.