CMO playbook & advanced tactics

The question that shapes a marketing organisation is rarely settled inside the marketing organisation. It gets settled in a board or CEO conversation, often in under half an hour: does this company need a CMO at all, and if so, what is attached to the title? Meta has not had a conventional brand CMO since Antonio Lucio left in 2020. Netflix ran for roughly three years with the seat empty, reinstated it, then churned through it. Mastercard went the other way and kept adding to it, communications first and later an operating business. Three answers, three very different bills. Your job as the incumbent, or the candidate, is to know which bill you are signing.

What the board is actually arbitrating

There are four moves available, and each one buys something real while shifting a cost somewhere less visible. Dissolve the title and fold the work into growth, revenue or public affairs. Split it, usually brand on one side and performance on the other. Expand it by attaching communications, customer experiencecustomer experienceThe overall perception a customer forms of your brand across every interaction, from first touch to post-purchase support.View full definition →, data or a P&L line. Or fuse it with an adjacent discipline so that one leader owns both marketing and the measurement apparatus that judges it. All four assume the commercial remit the foundations lesson sets out and the cadence and decision rights the frameworks lesson installs; what changes here is who holds them, and what happens to the business when the holder is a committee.

Dissolution: cheap on the org chart, expensive in the hiring market

2019 was the year the pattern went public. Uber folded its CMO role into a marketing and public affairs remit. Johnson & Johnson dispersed its global CMO responsibilities after Alison Lewis left. CocaCocaCustomer Acquisition Cost: total sales and marketing spend divided by the number of new customers acquired over the same period.View full definition →-Cola had already replaced its global CMO with a chief growth officer in 2017, and reinstated the CMO title in 2019 when it turned out that growth strategy and brand stewardship were not the same job. The pattern of the reversal matters more than the individual cases.

Dissolution reads as neutral on an org chart and lands as a downgrade in three places. First, the candidate market: senior operators price an SVP-reporting-to-CROCROConversion Rate Optimization (CRO) is the systematic practice of increasing the percentage of users who complete a desired action, using data, testing, and user research.View full definition → role differently from a C-suite seat, and you lose the top quartile of the pool before you start interviewing. Second, internal succession: marketing becomes a function people leave in order to get promoted, so your VPVPA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition → bench thins out over two or three years. Third, and this is the one that shows up in the numbers, nobody owns blended acquisition cost. Brand sits with the CEO or the founder, performance sits with revenue, and both sides can produce a dashboard that exonerates them while the mix drifts toward whatever is easiest to attribute this quarter.

Dissolution is defensible in exactly one situation: a founder who genuinely is the brand and intends to spend their calendar on it. The cost arrives later, when that founder leaves or the company enters a category where their personal credibility does not transfer.

The split, and where the seam breaks

Splitting brand from demand looks like specialisation and behaves like an unrefereed dispute. The observable symptom is specific: brand trackingbrand trackingRegular measurement of brand health metrics (awareness, image, preference, and purchase intent) over time, so shifts can be detected and linked to marketing activity.View full definition → flat or drifting down for three or four consecutive quarters while cost per acquisitioncost per acquisitionCost Per Acquisition: the total cost to generate one customer or conversion, computed by dividing total spend by the number of acquisitions.View full definition → climbs, and neither leader accepts the trade because neither is measured on it. Paid spend quietly becomes the patch for awareness the brand side was supposed to build, and the patch compounds, because every quarter of weak brand makes the next quarter of media more expensive.

The split does work when the two constituencies are actually different: a consumer brand talking to households, and a business-to-business sales organisation talking to advertisers or channel partners. Two audiences, two P&Ls, two leaders. What kills companies is splitting one audience across two leaders and calling it focus.

Expansion: what you attach, and what you can no longer decline

Attach communications and you get message discipline across earned and owned, plus a seat in every crisis. Attach customer experience and you get the post-purchase economics that most marketing scorecards ignore. Attach a P&L and you stop being a service function.

The cost is calendar and successor risk. A chief marketing and communications officer with a business unit attached will spend a meaningful share of any given quarter on matters with no demand-generation upside at all, which is fine while the incumbent is strong and brutal when they leave. The larger the bundle, the smaller the pool of humans who can carry all of it, so expansion tends to force an unbundling on exit. Plan the unbundling before you need it, or your search firm will plan it for you at 25% to 33% of first-year compensation.

Fusion: marketing plus analytics, marketing plus product

Fusing marketing with the measurement function ends the attributionattributionA framework for assigning credit to the touchpoints that contributed to a conversion, so you can measure which channels and interactions actually drive results.View full definition → argument by putting both sides of it under one leader. That is a genuine gain: no more three-week standoff between a marketing mix modelmarketing mix modelA statistical approach that estimates how each marketing channel and other factors drive sales, guiding budget allocation.View full definition → and a last-click dashboard. It also removes the internal check on marketing's own numbers, and it tends to select for leaders who are strong on growth mechanics and thin on brand construction. The second-order effect is that brand becomes an orphan: real, expensive, and owned by nobody senior enough to defend it in a budget cut.

Real-world case 1: Meta's fusion of marketing and analytics

Meta (which sells advertising, the thing this section is about, so read its structural choices as partly self-interested) had a traditional CMO in Antonio Lucio from 2018 to 2020. It did not replace him with another brand marketer. In early 2021 Alex Schultz, a long-time growth and analytics leader, took the title of CMO and VP of Analytics. Marketing and the numbers that judge marketing sit in one org.

What that buys: measurement disputes are settled internally, fast, and marketing speaks the same language as a company whose culture is quantitative to the point of ruthlessness. Meta's marketing and sales expense runs at roughly $12 billion a year, so even small allocation improvements are large absolute sums.

What it costs: Meta's hardest brand problems between 2021 and 2023, trust, regulation, teen safety, were not problems a growth-analytics org is built to solve, and they ran alongside roughly 21,000 job cuts across 2022 and 2023. The structural read is that Meta decided its reputation was a policy and product matter rather than a marketing one. Defensible for Meta. Fatal for a company whose pricing power depends on how people feel about the logo.

Real-world case 2: Netflix, empty seat then revolving door

Netflix operated for roughly three years after Kelly Bennett's 2017 departure without a CMO, pushing marketing decisions into the content organisation where the titles were made. That works when the product is the marketing and every title launch is its own campaign. It stops working when the business model changes.

Netflix then reinstated the title and cycled through holders in short order, on a marketing expense line of around $2.5 billion a year. The November 2022 launch of the ad-supported tier created a second constituency that the content-adjacent structure had never served: advertisers, agencies and a sales narrative, eventually organised under a separate advertising leadership. The lesson for the arbitration is that structure has a shelf life tied to the revenue model. A dissolved or distributed marketing function that fits a subscription business does not fit a two-sided one, and the gap shows up as a scramble to hire, not as a strategy.

Real-world case 3: Mastercard and the expanded seat

Mastercard went the opposite direction from Meta and Netflix. Raja Rajamannar has held the marketing brief for more than a decade, as chief marketing and communications officer, and for a period also ran the company's healthcare business. Communications folded in rather than sitting adjacent. On roughly $25 billion of annual revenue, the same leader signed off on the 2019 decisions to drop the wordmark from the logo and to build a sonic identity, moves that only make sense if one person holds brand, comms and enough tenure to be judged on the five-year outcome rather than the next quarter.

Average CMO tenure sits around four years, the shortest in the C-suite, which is what makes tenure itself a structural asset: brand decisions with long payback windows only get made by people who expect to still be there. The cost sits in the succession file. A bundle that large, held that long, cannot be handed over intact, and the honest version of this playbook is that Mastercard's next marketing structure will look different from the one Rajamannar built, whatever the title says.

CMO action items

• Write down which of the four moves your current structure represents, then name the cost you are currently paying for it. If you cannot name a cost, you have not understood the structure.
• Find the seam. Identify the one number nobody owns end to end (usually blended acquisition cost or net revenue retention) and force a written owner within one quarter, referencing the team operating model your organisation already runs on rather than redesigning it.
• Build the unbundling plan for your own seat: which two pieces of your remit would you separate if you left in six months, and who internally could take each. Give it to your CEO before they need it.
• Before accepting an expanded title, ask what percentage of the last four quarters the incumbent spent on crisis and stakeholder work. That is the share of your calendar you will not spend on demand.

Common mistakes that kill results

• Treating dissolution as a cost decision. The org chart saving is small; the pool of senior candidates you can no longer attract, and the internal bench that stops forming, are the real numbers, and they surface two years later.
• Splitting brand from performance across one audience. Two leaders, one customer, no referee: media spend becomes the patch for awareness nobody is building, and the patch gets more expensive every quarter.
• Accepting an expansion without moving anything out. Communications, CX and a business unit added to an unchanged marketing remit produces a leader who attends everything and owns nothing.

Key takeaways

• The role itself is an arbitration with four options: dissolve, split, expand, fuse. Each one is right somewhere and each one sends a bill to a part of the business that was not in the meeting.
• Structure has a shelf life set by the revenue model. Netflix's distributed marketing suited a pure subscription business and broke when advertising arrived.
• Fusion with analytics ends the measurement argument and orphans brand. Meta made that trade knowingly, on roughly $12 billion of annual marketing and sales spend.
• Expansion buys coherence and tenure, as Mastercard's decade-long bundle of marketing and communications shows, and the price is a successor problem that only gets harder the longer it works.
• Whatever the title, insist on one written owner for the number that spans the seam. Structures that leave blended acquisition cost or net revenue retention unowned fail the same way every time.