CMO playbook & advanced tactics for board & CFO communication

The moment a company is large enough for an activist fund to build a position in it, the marketing budget stops being an internal negotiation and becomes a public number. In 2017 Trian Partners took a stake of roughly $3.5bn in Procter & Gamble and ran a proxy fight that cost both sides tens of millions. Part of Trian's case was that P&G's spending, marketing very much included, was not buying growth. The vote was close enough to require a recount, and P&G seated Nelson Peltz on its board. Every marketing commitment made in Cincinnati afterwards was made in front of someone paid to test it.

That is the altitude this lesson works at. Who sits on the board and how a CFO screens a proposal is settled ground by now, as is the method for building the numbers. What follows is what happens when those numbers leave the building.

Sub-Concept 1: A published number is a floor, not a forecast

An internal forecast can be revised in a Tuesday meeting. A number spoken on an earnings call, printed in an investor day deck, or quoted to a trade title is one you own for several quarters, and it will be read back to you by someone who kept the transcript. Three tests before you let a marketing figure go public: can you produce it the same way in twelve months from the same source data; does it survive a media market that moves 20% against you; and can you explain a miss without blaming your own model.

Efficiency commitments are the most dangerous kind because they are asymmetric. If you announce that you removed $100m of waste and growth held, you have also announced that the waste existed. The next question is where the rest of it is, and it arrives every quarter after that, forever. There is no version of that story where you get credit once and move on.

Sub-Concept 2: Arbitration under a margin mandate

When a board hands down an operating margin target, the CMO does not get to argue about whether. You choose sequence, and sequence is the whole job. Most marketing lines carry an exit cost: contracts, notice periods, headcount conversations, sponsorship termination clauses. The one line that usually carries none of that is brand and upper funnel, so it goes first. Its payback is also the longest, so nothing visibly breaks for two or three quarters. Then pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.View full definition → coverage thins in Q4 and nobody in the room connects it to a decision made in Q1.

The defence is to date the warning. Put the expected lag in writing, in the same document as the cut, naming the quarter you expect the effect to land and the leading indicator that will show it first. Sent after the fact, that same analysis is an excuse. Sent before, it makes you the only person in the room who forecast the consequence.

Sub-Concept 3: The opposite commitment, promising to need less

Wise went public through a direct listing in London in 2021 and tells investors something unusual: it holds its underlying profit before tax margin inside a stated band and pushes anything above that band back into lower prices for customers. The majority of its new customers arrive through word of mouth rather than paid acquisitionpaid acquisitionVisitors arriving via paid ads or sponsored placements, where you pay a platform to display your message rather than earning visits organically.View full definition →. Marketing's mandate under that model is not growth at any CACCACCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition →, it is keeping acquisition cost from creeping while volume compounds.

The cost of that commitment is that you have no lever to buy a quarter. If referral rates slip, you cannot bid your way back to the number without breaking a public promise about price. It only works where the product mechanically generates referral through repeat, low-friction use. A low-frequency, high-consideration purchase cannot carry a Wise-style promise, and CMOs who borrow the framing without the underlying purchase behaviour end up defending a structurally impossible CAC.

Sub-Concept 4: The over-promise ledger

Four ways CMOs commit to things they cannot deliver:

• Guiding on blended CAC while planning a shift into brand. Blended CAC rises first when money moves upstream. You have guided yourself into explaining a deteriorating number that is the plan working exactly as designed.
• Promising a payback period that depends on gross margingross marginGross margin is the share of revenue left after subtracting the direct cost of producing goods or services, expressed as a percentage of revenue.View full definition →. One pricing decision or a heavier discount quarter moves payback, and finance made that decision, not you. You still own the miss.
• Extrapolating a full-year commitment from nine months in a seasonally skewed category.
• Attaching your name to anything competitor-dependent: share of voiceshare of voiceYour brand's share of total advertising or conversation volume in your category, measured against competitors over a defined period.View full definition →, category leadership, relative brand ranking. A rival's budget decision can break your commitment.

Real-World Cases

Case 1: P&G's Marc Pritchard used a January 2017 IAB speech to demand viewability standards, MRC-accredited measurement and contract transparency from the digital supply chain, then acted on it, cutting more than $100m of digital spend in a single quarter with no detectable impact on growth. By 2018 P&G reported roughly $750m taken out of agency and production costs, with the agency roster cut by more than half. What made this survivable at board level was that each cut was announced with a mechanism attached: fraud, non-viewable inventory, duplicated agency layers. The savings read as waste removal, not capacity removal. That distinction bought the option to spend again later, and P&G did increase advertising investment in the years that followed while arguing that brand plus product superiority drove volume. Had Pritchard sold 2017 as "marketing is over-funded", the later increase would have been a public reversal instead of a continuation.

Case 2: In January 2023, with Elliott Management, Starboard Value, ValueAct and Inclusive Capital on the register, Salesforce cut about 10% of its workforce and committed publicly to a 30% non-GAAPGAAPThe standard set of accounting rules companies follow to prepare consistent, comparable financial statements, dominant in US reporting.View full definition → operating margin. It reached that level ahead of the promised timetable. Note the conflict: Salesforce sells the marketing and analytics software companies use to make exactly these arguments, so its own internal discipline doubles as a sales asset. The instructive part for a CMO is what happened to the baseline. Hitting a margin commitment early does not return the money. The reduced spend level becomes the new normal against which every future request is judged, and the marketing programmes that survived did so because they could be defended per pipeline dollar rather than per impressionimpressionThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.View full definition →. If you are going to overdeliver on a cost promise, decide in advance what you will ask for in exchange, and ask while the goodwill is fresh.

CMO Action Items

• Write a one-page policy naming the two or three marketing figures you are willing to let leave the building, with the definitional rule for each. The test is whether an analyst who joins in a year can reproduce the number from the same data without calling you.
• Before you accept any cut, send the CFO and CEO a note naming the quarter the effect lands and the leading indicator that will show it first. Same week as the decision, not later.
• Model what blended CAC does in the two quarters after you move a fifth of spend upstream, then guide on a metric that does not invert during the transition.
• Answer this before the board does: if the company hits its margin target a year early, what do you ask for, and what evidence do you need to have collected by then to ask for it?

Common Mistakes That Kill Results

Mistake 1: Guiding on a number you do not control. Payback depends on gross margin, and gross margin depends on pricing and discounting decisions made elsewhere. If you commit to a payback figure in front of investors, you have accepted responsibility for a variable the commercial team can move without telling you. Commit to the spend and the pipeline; let finance own the margin assumption in writing.

Mistake 2: Cutting the longest-payback line because it is the only one without a contract. Brand and upper-funnelfunnelThe customer journey from awareness to purchase, typically Awareness, Interest, Consideration, Decision, Action, with prospects narrowing at each stage.View full definition → spend is easy to defer and expensive to restart, and the damage surfaces two or three quarters later when nobody is looking for a cause. Cut something with a visible short-term consequence instead, so the trade-off is discussed while the decision is still reversible.

Mistake 3: Changing the attribution model in the middle of a public commitment. Re-baselining makes every previous quarter unverifiable, and a board that cannot reconcile this year to last year concludes you are hiding something, whether or not you are. If the model genuinely has to change, run both in parallel for two quarters and publish both, including the quarters where the old model flattered you.