Skip to main content

Transform Before You Exit

A big transformation only makes sense if you want a generational exit, and you have to do it before you sell, because the buyer pays for the value you have already proven, not the potential you describe.

There are two honest answers to the question "should I invest heavily in AI transformation before I sell?" The first is no, if you are looking for a quiet, fair-market exit and you want to be done. The second is yes, if you want a generational outcome, the kind of number that changes your family's trajectory. This page is about the second answer, and about the timing trap that quietly destroys most of the upside.

Transformation Is a Bet, So Name the Prize​

A serious transformation is real money, real disruption, and real management attention. You should not do it casually, and you should not do it because a vendor told you to. You should do it for one reason: you have decided you want the maximum outcome, and you have accepted that the maximum outcome requires the maximum preparation.

This is the same logic that underlies the argument that grow or sell is the same decision. The work that produces a generational exit and the work that produces a dramatically better company to keep are the same work. So the bet is not really "spend on AI to sell better." The bet is "build a better company," and the exit is one of several ways that bet pays off. That framing should make the decision easier, because the downside path still leaves you with a stronger business.

But if the prize is a generational exit specifically, then timing becomes the whole game.

Buyers Pay for Proof, Not Potential​

Here is the trap. Owners imagine they can describe the AI upside to a buyer, point at the roadmap, and get paid for it. They cannot. A buyer's entire job is to pay for what is real and capture what is merely possible for themselves.

When you walk into a sale and say "with AI, this business could do far more," you have just handed the buyer their thesis for free. You have told them exactly where the upside is, and you have proven none of it. So they price the business on today's numbers, pocket the roadmap you described, and execute it after closing on their own dime. This is the heart of quantifying the acquirer's upside: unrealized potential is the buyer's profit, not yours.

Snider makes the underlying point in Walking to Destiny: you are not selling a story, you are selling a de-risked, transferable stream of cash flow, and the more of the value you have already built and proven, the more of it shows up in the price. McDannell is blunter in Get Acquired, that a buyer is purchasing an asset that already works, not a promise that it might. Potential is a discount. Proof is a premium.

Proven Value Compounds Into the Multiple​

The reason this matters so much is that proven AI value does not just add to your earnings. It moves your multiple, and the multiple is where generational outcomes come from. A business that has already demonstrated AI-driven margin expansion and a widening operational edge gets read as the kind of asset a buyer competes to own, which is the whole argument in AI now decides your multiple.

So the same transformation has wildly different value depending on when you do it. Done before the sale, with a year or two of real numbers behind it, it lifts both your earnings and the multiple applied to those earnings, and the two multiply together. Described before the sale, with nothing proven, it lifts nothing and simply educates your buyer. The dollars are the same. The outcome is not.

The Time You Need Is the Time You Are Spending​

The obvious objection is that transformation takes time you may not have. That is exactly the point. Proof requires runway. You need enough quarters of clean, attributable results that a diligence team cannot wave them away.

This is why the worst time to start is when you have already decided to sell. By then the runway is gone, and you are forced to sell potential instead of proof. The owners who win the generational outcome started the transformation while they still thought of themselves as operators, not sellers, which is just another expression of the idea that an exit is a posture, not an event. Burlingham's Finish Big makes the same case from the human side, that the owners who finish big are almost always the ones who prepared for years rather than months.

If you want the generational number, the move is not to describe a transformed company to a buyer. It is to become one before they ever see your data room.

Further reading​

Sources: Snider, Walking to Destiny; McDannell, Get Acquired; Burlingham, Finish Big; EisnerAmper 2025 on AI and business value.