Budget negotiation: frameworks & methodology

Friday deadline, blank sheet: build next year's marketing number from zero and be ready to defend every line to someone whose first question is what breaks if that line goes to zero. This lesson assumes you already know the cost lines the foundations lesson lays out and who owns each one. What follows is the arithmetic: how to build the number upward, how to tell an efficiency gain from an effectiveness gain, and how to hold a specific line item when a 15% cut lands on the table.

Efficiency and effectiveness are different sums

Efficiency is cost per unit of delivery: CPMCPMCost Per Mille: the cost to deliver 1,000 ad impressions. A pricing and benchmarking metric for awareness campaigns where reach matters more than clicks.View full definition →, cost per clickcost per clickCost Per Click (CPC) is the average amount you pay each time someone clicks your ad. It is a core pricing metric for paid search and social advertising.View full definition →, cost per opportunity, cost per asset produced. Effectiveness is profit generated per unit of spend. They move independently, and mixing them up is the fastest way to lose control of a negotiation. A 30% reduction in CPM tells you nothing about revenue, and a CFO who has been handed efficiency metrics for three years will assume you have no effectiveness data at all.

The sum that decides the negotiation is marginal, not average. Take a paid search line running at $5M a year and returning $15M of incremental gross profit. Average return, 3:1, looks fine. Fit a response curve and the picture changes: the first $1M returns roughly $6M, the fifth $1M returns about $0.9M. That last million destroys value. Now look at a retail media line at $1M returning $3M, still on the steep part of its curve, where the next $1M would return something like $2.6M. Moving the saturated million buys around $1.7M of extra gross profit at zero incremental budget. That is the calculation to walk in with, because it converts "how much do you need" into "which million are we discussing".

Two conditions have to hold before that sum is honest. You need some incrementalityincrementalityThe share of results (sales, conversions, revenue) that only happened because of a marketing action, not what would have occurred anyway.View full definition → read (geogeoThe practice of making your brand and content visible and citable inside AI-generated answers from tools like ChatGPT, Gemini and Perplexity.View full definition → holdouts, on/off tests, matched markets), and a curve fitted on more than a single year. Without both you have attribution, not marginal economics, and a good CFO will find the difference.

The split between long-term brand work and short-term activation is the other structural number. Binet and Field's IPA analysis puts the profit-optimal balance around 60% brand and 40% activation, shifting toward activation in high-consideration and online-only categories and toward brand in low-involvement FMCG. Treat it as a prior you can defend, not a rule. Its practical value is that it gives you a reason to keep brand money out of a marginal-return calculation that can only see 90 days.

Framework 1: the investment portfolio model

Divide the request into tiers before anyone else divides it for you:

• Tier 1: proven spend with documented return above the company's hurdle ratehurdle rateThe blended return a company must earn to satisfy everyone who funds it (lenders and shareholders). It sets the bar every investment has to clear.View full definition →. Defended first, cut last.
• Tier 2: growth bets with a payback projection built from comparable programmes.
• Tier 3: capped experiments with pre-written test criteria and a stop date.

AB InBev runs a version of this at portfolio level rather than channel level: global brands (Budweiser, Corona, Stella Artois), multi-country brands, then local champions, each with a different investment logic and a different expectation of what the money returns. The point is the same. When leadership asks for 15%, you hand over the Tier 3 list and state exactly which questions the company stops being able to answer next year.

Framework 2: zero-based build-up

Do not start from last year's number. Start from the activity, in this order:

1. List the jobs the business needs done next year, taken from the commercial plan: launch in three markets, hold share in the category leader's home market, recover lost distribution in a fourth.
2. Price each job in physical units: markets, weeks on air, assets required, 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 → needed against the named competitor.
3. Attach unit costs from actual invoices, not indexed estimates.
4. Sum, then compare with last year. The gap is your negotiation, and you now know which job dies if the gap closes.

AB InBev built its cost culture on zero-based budgetingzero-based budgetingA budgeting method where every expense is justified from zero each cycle, rather than carrying last year's spend forward.View full definition → inherited from 3G Capital, with every cost package owned by a named person who rebuilds it each cycle and defends it in a monthly review. Sales and marketing there runs in the high single-digit billions of dollars. What makes it work is not the zero. It is that each owner knows the driver sitting under each line (cases sold, media weeks, activations per outlet), so a challenge gets answered with a quantity rather than an opinion.

The twist for a CMO: apply the method fully to lines with 18 months or more of performance data, and carve out brand investment with a stated floor. Zero-based cycles are annual, so anything with a payback beyond twelve months loses by default unless you have written its protection into the process in advance. A floor expressed in finance's units works best: a percentage of net revenue, or excess share of voice against your main competitor, using the IPA finding that roughly 10 points of ESOV associates with about half a point of annual market sharemarket shareThe percentage of total industry sales your company captures in a given period. It measures competitive position relative to rivals in a defined market.View full definition → growth.

Framework 3: defending a line, then modelling the cut

For every line above a material threshold, prepare four things: the driver that moves it, the zero case with your most recent evidence for it, the cliff, and the trade you would accept instead.

The cliff is the least prepared and the most useful. Most lines are not linear and the CFO assumes they are. Cut a market launch's media by 40% and you do not get 60% of the result: you get a burst below effective frequency, subcritical share of voice against an incumbent, and a launch that has spent the money without buying the entry. Saying that in units, with the frequency and share-of-voice numbers attached, is what stops a proportional cut better than any argument about brand.

Then submit three numbers rather than one: base, growth and constrained. Write the constrained case yourself, at the figure the CFO is actually working with (ask privately, before the meeting). It should carry a non-linear response, a sensitivity table showing revenue delta per $250K added or removed, and a recovery cost: what rebuilding the abandoned position will cost later, and how many quarters it takes. Share bought back is almost always dearer than share held.

Real-world cases with numbers

Unilever moved to zero-based budgeting from 2016 under CFO Graeme Pitkethly, cut the number of ads it produced by around 30% and roughly halved its agency roster, and reported about €500 million of savings in brand and marketing investment in 2017. Most of that came out of the fee and production structures the agency management lesson describes, not out of working media. That distinction is the whole defence available to you: if you can show the cut lands on non-working cost, you can accept a savings target without accepting a loss of effect. If you cannot show it, every euro looks alike to the person holding the pen.

Kraft Heinz is the counter-example, and the reason to name a floor. After the 2015 merger the company ran zero-based budgeting across the combined business against a cost synergy target of roughly $1.5 billion, and margins rose quickly. In February 2019 it took a $15.4 billion write-down on the Kraft and Oscar Mayer brands, cut its dividend and disclosed an SEC subpoena related to procurement accounting. Miguel Patricio, arriving from AB InBev later that year, said publicly that the business had underinvested in its brands and raised marketing spend. Put the two figures side by side: impairment measured in the tens of billions against savings measured in the low billions. The method found genuine waste, kept cutting into working spend, and the balance sheet recorded the difference three years later.

CMO action items

• Fit a response curve for your two largest channels before planning season and state average and marginal return separately. If you can only produce the average, say so first; being caught later costs more.
• Write the constrained scenario yourself, at the number finance is actually modelling, and include the recovery cost of what it gives up.
• Set a brand floor in a unit finance accepts (percentage of net revenue, or ESOV against a named competitor) and get it agreed before the cycle starts, not during it.
• Rebuild your three largest lines from unit drivers, so any challenge gets a quantity back within the meeting.

Common mistakes that kill budget negotiations

• Bringing average returns to a marginal question. "This channel runs at 4:1" invites a cut. "The next $500K in this channel runs at 1.1:1, the next $500K in that one at 2.4:1" directs it.
• Anchoring on last year. Reference last year's number and you have conceded that the only live question is the increment.
• Accepting a cut spread evenly across lines. Proportional cuts push every line toward its cliff at once. Fully funding fewer jobs beats half-funding all of them.
• Selling efficiency wins as effectiveness. Tell the CFO you took 22% out of CPMs and next year's opening ask is another 22%. Efficiency gains are one-off; treated as a trend, they compound against you.
• Going to finance before the commercial owner whose plan your lines pay for has said, in writing, that the plan needs them.