PlaybookMarch 29, 20269 min read

Defending a GEO score in front of your CFO: a script

Your CFO doesn't care about citations. They care about whether the line item produced a return. Here's the seven-minute conversation that turns 'AI visibility is up' into 'this is the budget we should keep, and here's why.'

Most CMOs lose the GEO budget conversation in the same place every year: about ninety seconds in, when the CFO says “but what does it actually do for revenue.” The answer is rarely good. It’s usually a long set of qualitative claims about authority and category leadership, delivered with a little too much intensity, followed by a promise to do better next quarter on the attribution side.

That conversation is losable not because the underlying investment is bad — it isn’t — but because the CMO walked in with the wrong artifact. This post is the script we’d hand a CMO heading into that meeting. It’s a seven-minute conversation, not a fifty-slide deck. It’s designed to survive a hostile finance audience and to leave the CFO with a number they can put in the model.

The frame: what a CFO actually buys

A CFO is not buying “AI visibility.” A CFO is buying one of three things from your line item, and your job is to know which one before the meeting starts. The three are:

Risk reduction.“If we don’t spend this, our pipeline collapses in 24 months as buyer discovery shifts to AI.” This is the easiest frame to defend in a downturn because it positions the spend as insurance.
Pipeline efficiency.“Buyers who arrive having already seen us cited by AI close at 1.4x the rate of cold pipeline, and demand 30% less SDR time.” This is the easiest frame to defend in a growth quarter because it positions the spend as a margin lever.
Strategic positioning.“The two companies that win the AI-citation race in our category will define the category. We’re positioned to be one of them.” This is the hardest frame to defend in a finance meeting and the easiest to defend in front of the Board. Use it sparingly with the CFO and load up on it with the Board chair.

Pick one frame. Don’t use all three. The CMO who rotates through all three frames in the same meeting is the CMO whose budget gets cut, because the rotation reads as uncertainty about what the spend is actually for.

The seven-minute script

Verbatim, with stage directions. Adjust the numbers to your own. We’ll use the pipeline-efficiency frame as the worked example because it’s the hardest to fake.

Minutes 1–2: anchor on the metric they already trust

“You and I look at pipeline conversion every month. I’m here to talk about a specific source of pipeline — buyers who arrive already convinced because the AI tool they trust recommended us before we ever spoke. That source is now [X]% of inbound. Last year it was [Y]%. It’s the highest-margin source we have because the sales-cycle work is mostly already done.”

You haven’t mentioned GEO, citations, or AI visibility yet. You’ve anchored the conversation on a metric the CFO already cares about and brought a new fact about it. The CFO is now leaning forward, not crossing their arms.

Minutes 3–4: make the causal claim, defended

“The reason that source is growing is that we’ve been investing in being the brand AI assistants recommend. When ChatGPT or Gemini gets asked a category question in our space, we’re now the explicit recommendation [X]% of the time. Two quarters ago that number was [Y]%. The line item that moved it is the GEO program. The line item is [$Z] per month. The pipeline attributable to it is [$N] in closed-won this year so far.”

Three sentences, three numbers, one causal claim. Notice what isn’t in there: no “mention rate,” no sentiment score, no citation count. Those are technical diagnostics for your team, not boardroom artifacts. The only number that survives the elevator from the GEO tool to the CFO is the recommendation rate, because it’s the only one that maps cleanly to a buyer behavior the CFO already understands.

Minute 5: pre-empt the obvious objection

“The objection I’d expect from you is that we’re attributing pipeline that would have come in anyway. Two responses. One: the deals tied to this source consistently arrive at higher buyer-stage scores than our cold pipeline, which we know because [your sales team tracks this]. Two: when we paused content investment in the program for six weeks last fall to test, the recommendation rate stalled and inbound from this source dropped by [%]. Both data points point in the same direction.”

If you don’t have a real pause-test or a real buyer-stage score difference to cite, don’t pretend you do. CFOs catch this immediately. Substitute the nearest honest equivalent: a competitor who stopped investing and saw their citation rate collapse, a category prompt where you can show a clear before/after on a specific content investment, or simply concede the attribution gap and explain how you’re reducing it.

Minute 6: hand them the artifact

“I’m sending you a one-pager. It has the recommendation rate trend, the inbound source mix, and the line-item cost. The right unit-economics question is whether the program returns more in pipeline efficiency than it costs. The current answer is [Nx]. I’d like to keep the program at current investment, with one contingency: if the recommendation rate drops below [%] for two consecutive months, we’ll cut investment by a third and re-evaluate.”

The contingency is the most important sentence in the whole script. It tells the CFO you’re not asking for a blank check — you’ve already pre-committed to cutting the budget yourself if the metric stops moving. That posture earns more program continuity than any ROI claim.

Minute 7: leave

Genuinely. Stop talking. Don’t fill the silence with more justification. CMOs lose budget conversations in the last ninety seconds of the meeting, after they’ve already won, because they keep talking past the close. Hand over the one-pager. Ask if there’s anything else. If not, leave.

Six elements, in this order, on a single page in plain type. Charts only if they’re necessary.

The recommendation rate (current and 12-month trend). Two engines, charted separately, no composite. If the trend is up, the chart sells the line item by itself.
The inbound mix.What percentage of inbound pipeline arrives “AI-influenced” (define the operational rule and stick to it). Show how the mix has shifted over 12 months.
The unit economics. Program cost per quarter divided by attributable closed-won pipeline. One number.
The contingency.The trigger condition under which you’ll cut your own budget. Make it specific and measurable.
One concrete example. One named deal (or anonymized if necessary) where the buyer arrived having read about you in an AI-assistant answer. The CFO will remember the story long after they forget the numbers.
What you’re not asking for.One line that says “not requesting an increase” (or whatever the actual ask is). CFOs read for the ask first; making it explicit prevents them from inferring a bigger one.

The deeper point

The CMOs who keep their GEO budgets through the next downturn won’t be the ones with the prettiest dashboards. They’ll be the ones who learned to translate citation data into the financial vocabulary the CFO already speaks. That translation is uncomfortable, because it requires throwing away ninety percent of what the GEO tool tells you and surfacing only the one or two numbers that map to a P&L line. But it’s the translation that survives the budget cycle.

The good news is that the translation is doable. The bad news is that nobody else is going to do it for you. The GEO vendor will hand you a dashboard with twenty-eight metrics. Your CFO needs three. The work in between — the editorial decision about which three — is the actual marketing leadership job.

Written by The Enso team. Have a question or correction? Email us.