01There is no neutral
Founders like to believe there’s a safe middle position on AI: use it quietly, say little, wait for the noise to settle. In a sale process the middle doesn’t exist. An agency with no AI story is assigned one, and the assigned story is always the same: this business is a cost line someone else’s software is coming for. A diligence team will look at your delivery (production, reporting, media operations) and mark a meaningful share of it as automatable. On the businesses we assess, a third of delivery reading as automatable is not unusual.
Silence is a position. It’s just the worst one, because it hands the most important paragraph of your equity story to the person whose job is to pay you less.
Value already created but still priced against you: AI running inside your delivery while absent from your margin evidence, your offer, and your equity story. Latent, because the premium exists; it just hasn’t been made legible to the person paying for it.
02The two tests
Strip away the noise and a buyer asks exactly two questions about your AI. Is it in the P&L? Measurable delivery efficiency, hours per deliverable trending down, margin holding above the line quality buyers underwrite. Is it in the story? A service line clients pay for, revenue you can attribute, a narrative the equity case can carry without blushing. Everything else (the pilots, the demos, the tools slide in the capabilities deck) is theater, and buyers have seen a great deal of theater lately.
Two tests, two answers each. Four squares. Every agency in the market today is standing in one of them.
03Where most good agencies stand
Bottom-left is AI-Threatened: no evidence, no story, priced accordingly. Top-left is Vapor: a confident narrative with nothing underneath, which is worse than silence, because it survives the first meeting and dies in the second week of diligence, taking your credibility with it. Top-right is the premium: evidenced and told, underwritten without argument.
But the interesting square is bottom-right, because that’s where most genuinely good agencies are standing right now. AI is in the work: production accelerated, reporting automated, delivery costs quietly down. And none of it is in the offer, the numbers a buyer sees, or the story. The efficiency is real and invisible. That’s the latent discount: paying the threat price for a business that has already done the premium work.
The most expensive AI strategy in the market right now is having one and not saying so.
04What evidence looks like
Converting the square is mostly a legibility exercise, and legibility is boring on purpose. Delivery hours per deliverable, trended across eighteen months. Margin against the twenty-percent line that quality buyers use as shorthand for an operation that works. Revenue attributed to AI-native offerings: an audit product, an optimization retainer, anything a client pays for by name. Clients who have moved off an hours-based rate card and now buy solutions and outcomes, which is evidence the delivery model itself changed rather than merely got faster. A moat you can point at: a product or an AI-enabled service nobody else in your lane offers, which collapses the comp set to a market of one. And an exposure audit you ran on yourself, showing which parts of delivery are automatable and what you’re doing about it, before the buyer’s consultant shows you their version.
Evidence is believed precisely because it’s dull. A buyer who has sat through a dozen AI decks will pay for the one agency that shows up with a trend line instead.
05The widest swing on the scorecard
Here is why this perspective exists: of the seven levers a buyer underwrites, AI is the only one that moves in both directions at once. Unmanaged, it’s a discount: the threat paragraph, priced in turns. Evidenced and told, it’s a premium buyers currently pay up for, because every acquirer wants the AI-capable platform and few can find one. The distance between those two prices, on the same underlying business, is roughly three turns. No other lever swings that far on work that is largely already done.
The order matters: margin first: measure what’s already true. Offer second: productize what the measurement proves. Story last, and only what the first two can support. Story-first is how agencies end up in Vapor, and Vapor is the one square worse than saying nothing.
AI is a latent discount, and your highest-return premium. The work is mostly done. What’s missing is the proof, and proof is a project, not a gamble.
Next in PerspectivesSell who you are, not what you have. · Positioning