Marketing budget cuts playbook: what to cut first, what to protect and how to win budget back
Cut martech and agency scope before media, hold share of voice at or above market share, and price every cut for finance, with triggers to win budget back.
A marketing budget cuts playbook cuts in a set order: unused martech seats and overlapping tools first, then agency scope, low-return programs and media last. Hold share of voice where it already matches or beats market share, and price every cut for the chief financial officer (CFO). Budgets were already thin: Gartner's 2026 survey found the average at 7.8% of company revenue, up only 1.3% since 2025.
What to do in the week the marketing budget is cut
Cut in a set order, hold share of voice where it already matches or beats your market share, and put a price on every cut before it reaches the chief financial officer (CFO). That is the week's work. The rest of this piece explains each step.
Start from the baseline. Gartner's 2026 survey found the average budget at 7.8% of company revenue, and budgets had grown only 1.3% since 2025. A cut on top of that lands on spend that is already thin. Our budget benchmarks show where you sit.
Gartner, CMO Spend in 2026: Redefining Marketing Investment Under Constraint, 25 June 2026
Cut in this order
| Order | What goes | Why it sits here (our reading) |
|---|---|---|
| 1 | Unused or overlapping martech seats and usage contracts | Customers do not see it, and savings arrive within a billing cycle |
| 2 | Agency scope of work | Scope can shrink while the relationship stays intact |
| 3 | Low-return programs | They cost effort as well as cash |
| 4 | Media | The damage shows up in the market, and it shows up late |
For tools, Gartner's advice to CMOs was to monitor usage, renegotiate contracts and add controls on consumption pricing. It also told CMOs to eliminate low-value legacy spend and automate routine work.
Gartner, CMO Spend in 2026: Redefining Marketing Investment Under Constraint, 25 June 2026
Media goes last because of the evidence on downturn cuts. An IPA (Institute of Practitioners in Advertising) seminar held on 4 March 2008 concluded that cutting budget in a downturn will only help defend profits in the very short term. One case study shown there had a long-term payback over four times greater than the short-term.
Protect share of voice where it holds
The same seminar advised keeping share of voice (SOV) at or above SOM (share of market) during a downturn. It also found that brands which cut relative to competitors are at greater risk of share loss. Where your SOV meets or beats your SOM, treat that media line as protected. Find the savings in rows one to three.
Price every cut for the CFO
Every cut gets a price tag. As an example, not a benchmark: on a budget of $5,000,000, a 10% cut is $500,000. If 20% of that $500,000 comes from tools, tools deliver $100,000.
For each line, write four things:
- Amount saved and the quarter it lands
- Pipeline or revenue at risk, with your estimate method
- The leading indicator you will watch
- The trigger that would bring the spend back
That last line is how you win budget back later. Our guide to CMO, CEO and CFO alignment covers how to present it.
Where marketing budgets already stood before this cut
This cut did not land on a full budget. Gartner reported that marketing budgets have been on a plateau since 2022 and that the current average is 18% lower than the mean budget allocation just four years ago. In 2025 budgets sat flat at 7.7% of overall company revenue, the same share as the year before. The small 2026 rise covered earlier did not win that ground back.
Gartner, CMO Spend in 2026: Redefining Marketing Investment Under Constraint, 25 June 2026
Gartner, Gartner 2025 CMO Spend Survey Reveals Marketing Budgets Have Flatlined at 7.7% of Overall Company Revenue, 12 May 2025
The cut itself was flagged in advance.
One caveat for readers below the enterprise tier. The 2026 survey covered 401 CMOs and other marketing leaders in North America, the United Kingdom and Europe, mostly at companies with revenue over $1 billion. Compare your own ratio against our percent-of-revenue benchmarks before you quote it to a board.
What had already moved
| Measure | 2025 | 2026 |
|---|---|---|
| Paid media share of marketing budget | 30.6% | 31.4% |
| CMOs saying budget falls short of strategy | Fifty-nine percent | 56% |
Gartner said the 2026 paid media gain was funded by cuts to agencies, after thirty-nine percent of CMOs planned to cut back on agency budgets in 2025. In our reading, the agency line many CFOs look at first may have little slack left.
Gartner, CMO Spend in 2026: Redefining Marketing Investment Under Constraint, 25 June 2026
Gartner, Gartner 2025 CMO Spend Survey Reveals Marketing Budgets Have Flatlined at 7.7% of Overall Company Revenue, 12 May 2025
AI is the other pre-committed line. CMOs put 15.3% of budgets into AI, yet only 30% report mature AI readiness. The AI-ready group allocates 21.3% to AI and runs budgets of 8.9% of company revenue. See AI share of budget for more detail.
A worked example
Take an example budget of $5,000,000, not a benchmark. A 10% cut on $5,000,000 removes $500,000, a 15% cut removes $750,000, and a 20% cut removes $1,000,000. On that same $5,000,000, a 15.3% AI share comes to $765,000.
Gartner, CMO Spend in 2026: Redefining Marketing Investment Under Constraint, 25 June 2026
Put three lines on one page for the chief financial officer (CFO):
- Where your budget stood before 2022, set against today.
- What you already cut: agency, labor, retention.
- What is locked in: AI contracts, media commitments.
We think this framing matters. A cut of 10 to 20 percent falls on budgets that were never rebuilt, and the CFO should see that before the number is final.
What to cut first, in order
The order below is our method, not a survey finding. It runs from cuts that are cheap to undo toward cuts that are expensive to undo. The rule is simple: cut first what you can switch back on quickly, and cut last what took years to build.
Copy the table into your planning file and replace our readings with your own numbers.
| Cut, in order | Typical saving (our reading) | Risk (our reading) | Reversal cost (our reading) | Time to reverse (our reading) |
|---|---|---|---|---|
| 1. Unused martech seats and overlapping tools | Small to medium, recurring | Low, if login data backs it | Low: re-license | Days to weeks |
| 2. AI pilots with no path to scale | Medium | Low to medium | Low: restart from the documented pilot | Weeks |
| 3. Agency scope and retainer hours | Medium | Medium: slower, thinner output | Medium: rebriefing and onboarding | A quarter or more |
| 4. Events and programs with low return | Medium to large | Medium: lost field pipeline | Medium to high: deposits, lost slots | A full booking cycle |
| 5. Performance media | Large, fast | High: near-term pipeline drops | Medium: bidding and learning reset | Weeks to a quarter |
| 6. Brand media | Large | High: share loss shows late | High | A year or more |
How to apply each row
Row one is an audit, not a negotiation. Pull seat usage from each tool's admin panel and cancel anything unused before the next renewal date.
Row two depends on readiness. The previous section showed how much budget already goes to AI and how few organizations can scale it. Gartner adds that 70% of CMOs also acknowledge that their internal processes are not mature enough to implement and scale AI. We think a pilot with no owner, no data source and no production plan belongs in row two, and a pilot with all three belongs in the protected column.
Row three works best as a scope change, not a fee haggle. Remove deliverables, then hours. Use the agency contract checklist to confirm notice periods first.
Row four should be judged on pipeline per dollar, checked against the event budget benchmarks.
Rows five and six come last because their costs show up after the saving does. Check your current split against the brand versus performance benchmarks before touching either.
Worked example
Use these as example inputs, not a benchmark. A $5,000,000 budget cut by 10% means finding $500,000. If rows one and two yield 20% of the $500,000 target, that is $100,000, and the remaining $400,000 has to come from rows three onward. Work down the table until the target is met, and stop there. Every dollar found higher up is a dollar you do not take from media.
Should you cut advertising in a recession
The short answer from the Institute of Practitioners in Advertising (IPA) evidence: cutting looks cheap this quarter and costs more later. Three bodies of work point the same way.
What the evidence says
The 2008 IPA seminar, whose short-term profit finding we covered earlier, also concluded that "Ultimately the brand will emerge from the downturn weaker and much less profitable." It added that when rivals are cutting, the longer-term benefit of holding share of voice (SOV) at or above share of market (SOM) grows.
Going dark carries its own cost. Seminar data showed that brand usage and brand image suffered considerably when brands stopped spending for six months or more.
Peter Field later examined IPA data from around 50 case studies covering the 2008-9 recession. His reading was that the brands that held their nerve and their share of voice bounced back strongly when recovery came.
His report Winning or Losing? found that reducing advertising investment during recessionary periods is associated with weaker long-term performance and slower recovery.
commscouncil.nz, Advertising in a Recession
Exposure varies by category
The seminar data showed some categories carry more risk than others.
| More exposed to share loss when cutting | More resilient |
|---|---|
| Motor fuel | Luxury cars |
| Mineral water | Financial services |
| Apparel | Fragrances |
We think price-led buyers are a minority. Around 10% of consumers across categories were exclusively motivated by price, which is why the seminar saw reason to keep building preference through a downturn.
A worked example for the board
These are example inputs, not a benchmark. Take a brand with a 15% share of market and a 20% share of category media spend, a gap of 20 minus 15, or 5 points. A cut that drops its SOV to 10% turns the gap into 10 minus 15, or minus 5 points, which puts it on the wrong side of the share-of-voice rule set out earlier.
Run that calculation before the cut is final. Then put three numbers in front of the chief executive officer (CEO):
- Current SOV minus SOM.
- SOV minus SOM after the proposed cut, assuming competitors hold spend.
- The same figure if competitors cut by the same proportion.
If the third line still shows a positive gap, we think the cut is easier to defend. If the second line turns negative, the trade-off belongs on the board agenda as a market-share risk, not a budget line. Our guide to CMO, CEO and CFO alignment covers how to frame it.
Track brand usage and image monthly during any cut, using the measures in our brand tracking metrics guide, with share of search as a cheap early signal.
What excess share of voice is and how to calculate it
Share of voice (SOV) is a brand's slice of all the advertising in its category. The IPA (Institute of Practitioners in Advertising) seminar report on downturns defined it as a brand's share of total category communications expenditure. Share of market (SOM) is the brand's slice of category sales. Excess share of voice (ESOV) is the gap between the two. The same report called SOV minus SOM the key expenditure metric when it looked at the risk of losing share.
A worked example
These are round example inputs, not benchmarks.
| Step | What you calculate | Example value |
|---|---|---|
| 1 | Your SOM, from sales data | 10% |
| 2 | Your SOV, from your spend divided by estimated category spend | 15% |
| 3 | ESOV, SOV minus SOM | +5 points |
| 4 | Cash removed by a 20% cut to a 5000000 budget | 1000000 |
Take a brand with a 10% SOM and a 15% SOV: its ESOV is 15 minus 10, or +5 points. A 20% cut to a $5,000,000 budget removes $1,000,000. What that does to ESOV depends on rivals. If the cut pulls SOV down to 10% against a 10% SOM, ESOV falls to 10 minus 10, or zero. That is the floor the IPA guidance asks you to hold, as the previous section set out. Any deeper cut puts the brand below it.
Two practical points. Category spend is an estimate, so use the same source every quarter and track the trend rather than the decimal. And put the ESOV figure in the board pack next to the cut itself, so the trade-off is visible in one line.
Why rivals matter more than your own number
ESOV is relative. If every competitor cuts by the same 20%, your SOV does not move. The IPA report made the same point: if brands were to cut expenditure equally, then apart from some minor effects on the category as a whole (the most important of which might be an increase in price sensitivity), there would be little impact on brands individually. In practice that rarely happens: some brands maintain or even increase their expenditure, whilst others cut theirs.
That is the mechanism behind the larger gain when rivals cut. In our reading, a brand that holds spend while competitors pull back gains SOV without paying for it, because the category total shrinks around it.
The far end of the scale is a SOV of zero. Ehrenberg-Bass Institute research, based on a longitudinal analysis of 365 brands in 22 categories, found that brands that stopped advertising lost share year after year. Market share declines were more common and substantial among small brands and those losing share before advertising ceased. If your brand is the smaller player, treat ESOV as a harder floor. Pair it with brand tracking, and use share of search as a faster check between sales reads.
Thinkbox, Strategies for marketing through a downturn, 17 July 2023
When a cut to brand spend is defensible, and how fast the damage shows
Kantar names one exception to the case for holding brand spend. When demand fades for a product group, it warns, "your growing market share might not equal profitability." Its example: just before its downfall, Nokia's market share was still growing, but the whole category was gradually shrinking. Defending share in a disappearing market buys a bigger slice of less profit.
So the first test of a defensible cut is category size, not brand share. Pull three years of category volume and value. If both are falling for structural reasons, such as a substitute product or a change in how buyers buy, we think the cut becomes a portfolio decision. The money then belongs where demand is moving. If the category is only dipping with the economy, the exception does not apply.
Why the cut looks fine at first
Cuts flatter the next few quarters. Findings from an Institute of Practitioners in Advertising (IPA) seminar held that cutting budget in a downturn will only help defend profits in the very short term. The brand keeps living off past investment, which produces a "dangerously misleading increase in short-term profitability," and the longer-term harm will not be noticed at first. In one case shown, the long-term element of payback was over four times greater than the short-term.
Reversing course costs more than the saving. Kantar's modeling found that repair takes several more years and more media investment than was originally cut. A UK beverage brand that went dark for a year in one region declined by 2 percentage points of share there, and resuming spend the next year did not recover the lost market share.
What to watch so slippage shows before sales
| Reading | What a slip looks like | How we would use it (our reading) |
|---|---|---|
| Brand bonding or relationship score | Falls ahead of share; Millward Brown showed a strong correlation between market share and the level of 'bonding' | Report monthly from your brand tracking |
| Brand usage and image | The dark-period damage covered earlier | Compare cut and uncut regions |
| Share of search | Your branded searches fall as a share of category searches | Check weekly; it moves fastest |
| Word of mouth | A silent brand will very rapidly see this manifested in word-of-mouth | Track social and review sentiment |
| Category volume | Falling for several quarters | Tests whether the exception applies |
Agree trigger levels for each reading with the CFO before the cut takes effect, and put them in the monthly board pack.
How to show the CFO the expected cost of each cut
A chief financial officer (CFO) will accept a cut list faster when each line shows three things: what it saves, what it is expected to cost and what it would cost to undo. Our method puts those three figures on one page, one row per line item.
A worked example
Take an example budget of $5,000,000 and a 15% cut, which is $750,000. These are round inputs for illustration, not a benchmark. Suppose one line, sized at 10% of the $5,000,000 budget, is $500,000. Removing that line entirely leaves $250,000 of the $750,000 target to find elsewhere, usually across two or three smaller lines.
For each line, the team then fills in two estimates from its own evidence.
- Expected pipeline or revenue cost. Use incrementality tests where you have them and marketing mix modeling (MMM) where you do not. Ask the model for both the short-term effect and the effect over one to four years. Econometric work presented at a March 2008 conference of consultancies put most short-term payback estimates at around 50%, with longer-term payback usually considerably greater. A cost estimate built only on the short window will understate the loss.
- Reversal cost. This is what it would take to rebuild the line later: rehiring, recontracting and relaunching media at a heavier weight. An earlier section covered why repair takes more than the amount cut.
Show the CFO options for each line as well as outright deletion. The same consultants modeled two scenarios: a budget cut to zero for one year, and a budget halved for one year, each then returned to usual levels. Halving a line and stopping it are separate choices, and the CFO can price each one.
The one-page template
| Line item | Saving | Expected pipeline or revenue cost (from your model) | Reversal cost (our estimate) | When the cost shows |
|---|---|---|---|---|
| Line A (example) | $500,000 | Short-term and one-to-four-year figures from MMM | Cost to restart at prior weight | Short-term part within the quarter, long-term part over years |
| Lines B to D (example) | $250,000 combined | Incrementality result for each line | Rehiring or recontracting cost | As each model shows |
| Total | $750,000 | Sum of rows | Sum of rows | Both windows |
Close the page with the trade-off. The conference summary held that if a brand keeps share of voice at or above share of market, "the longer-term improvement in profitability is likely to greatly outweigh the short-term reduction" (report). We think the page persuades because it states that trade-off in the CFO's own units: dollars saved this year set against pipeline lost later.
The 90-day timeline and the conditions for getting budget back
The cut is the easy part. The harder part is setting, in the same quarter, the conditions under which money comes back. In our reading, when that happens, the CMO who leaves with a written restoration deal is in a better position than the one who leaves with only a smaller budget.
| Window | Owner | Output | Success test |
|---|---|---|---|
| Week 1 | CMO, with the finance business partner | Signed cut list, named protected lines, draft trigger memo | CFO has countersigned the trigger memo |
| Weeks 2 to 4 | Marketing operations lead | Baseline scorecard for every trigger metric | Each metric has a baseline, a data source and one owner |
| Weeks 5 to 12 | CMO, reporting to the CEO | Monthly scorecard, reallocation of remaining spend, first formal trigger review | A tripped trigger releases its tranche with no new debate |
Restoration triggers to put in writing
Agree three triggers with the CFO. Each one releases a named tranche.
- Share of voice: your share of voice (SOV) falls back below your share of market (SOM), measured the way the earlier section set out.
- Pipeline coverage: marketing-sourced pipeline coverage of next quarter's target falls under a ratio the two of you write down as a number.
- Revenue: revenue returns to plan for a set number of consecutive months.
Here is a worked example with round inputs, not a benchmark. A $5,000,000 budget is cut by 20 percent. The tranche held in reserve is 20 percent of $5,000,000, or $1,000,000. The memo can split it so that 10 percent of $5,000,000, or $500,000, returns on the SOV trigger. The other $500,000 returns on either the pipeline trigger or the revenue trigger.
Bring the evidence to the board with the memo. Ehrenberg-Bass analyzed 70 consumer goods brands across 20 years. When those brands stopped advertising for a year or more, sales fell 16% after one year, and 25% after two years on average. A Harvard Business Review article summarized by Thinkbox found that firms that maintain their marketing spend while reallocating it typically do better than firms that cut. In our reading, this supports making reallocation, and not only reduction, part of the 90-day plan.
Thinkbox, Strategies for marketing through a downturn, 17 July 2023
The monthly scorecard
Report these each month in the marketing scorecard template. Carry the same page into the board deck.
- SOV minus SOM, by core category
- Share of search against the two closest rivals
- Brand tracking scores on the protected lines
- Marketing-sourced pipeline and its coverage ratio
- Revenue against plan
- Spend against the revised budget, with the status of each trigger
Show each trigger as either met or not met. Do not use a judgment call. That status is what turns the conversation from a request for budget into the release of an agreed tranche.
What this page does not cover
This page does not cover headcount decisions, how to choose or build a marketing mix model, or how to evaluate specific vendors and agencies. The Gartner benchmarks it cites came mostly from companies with revenue over $1 billion, so compare them with your own ratio before quoting them. It is not financial or legal advice, including on agency contract terms.
Related on CMO Tribune: Marketing budget as a percentage of revenue in 2026, by sector and business model; AI marketing budget 2026: the 15.3% benchmark, where the money goes, and how to set your own AI line; Event marketing budget percentage: B2B benchmarks, their samples and their denominators.
Questions
What should I cut first in a marketing budget?
Start with unused or overlapping martech seats and tools, then AI pilots with no owner, data source or production plan, then agency scope and retainer hours, then low-return events and programs. Performance media comes next, and brand media goes last. The rule is to cut first what you can switch back on quickly and cut last what took years to build. Every dollar found higher up the list is a dollar you do not take from media.
Should you cut advertising during a recession?
A 2008 Institute of Practitioners in Advertising (IPA) seminar concluded that cutting budget in a downturn only helps defend profits in the very short term and that the brand emerges weaker and less profitable. Peter Field's review of around 50 IPA case studies from the 2008-9 recession found that brands that held their share of voice bounced back strongly when recovery came. The one exception Kantar named is a category that is shrinking for structural reasons, where defending share buys a bigger slice of less profit.
What is excess share of voice in marketing?
Excess share of voice (ESOV) is share of voice (SOV), a brand's share of total category advertising spend, minus share of market (SOM), its share of category sales. A brand with a 15% SOV and a 10% SOM has an ESOV of +5 points. The IPA seminar advised keeping SOV at or above SOM during a downturn, so an ESOV of zero is the floor to hold. Because ESOV is relative, what rivals spend matters as much as your own cut.
How do I explain the impact of a marketing budget cut to the CFO?
Give the chief financial officer (CFO) one page with one row per line item showing the saving, the expected pipeline or revenue cost and the cost to reverse it. Estimate the cost with incrementality tests or marketing mix modeling (MMM), and ask for both the short-term effect and the effect over one to four years, since longer-term payback was usually considerably greater than short-term. Offer options such as halving a line as well as stopping it, so the CFO can price each choice.
How long does it take for brand damage to show after cutting advertising?
The harm shows late, because the brand lives off past investment and short-term profit looks better at first. IPA seminar data showed brand usage and image suffered considerably when brands stopped spending for six months or more, and Ehrenberg-Bass found sales of 70 consumer goods brands fell 16% after one year dark and 25% after two. A UK beverage brand that went dark for a year in one region lost 2 points of share there and did not recover it when spending resumed. Watch share of search weekly and brand tracking monthly to see slippage before sales do.
How do I get marketing budget back after a cut?
Write restoration triggers into a memo that the chief financial officer (CFO) countersigns in the week of the cut, each releasing a named tranche. Typical triggers are share of voice falling back below share of market, marketing-sourced pipeline coverage dropping under an agreed ratio and revenue returning to plan for a set number of months. In a worked example, a 20% cut to a $5,000,000 budget holds $1,000,000 in reserve, with $500,000 returning on each of two triggers. Report each trigger monthly as met or not met, so a tripped trigger releases money with no new debate.
Sources
- CMO Spend in 2026: Redefining Marketing Investment Under Constraint, Gartner, 25 June 2026
- Advertising in a downturn: A report of key findings from an IPA seminar (March 2008), IPA
- Gartner 2025 CMO Spend Survey Reveals Marketing Budgets Have Flatlined at 7.7% of Overall Company Revenue, Gartner, 12 May 2025
- Gartner 2026 CMO Spend Survey Finds CMOs Allocate 15.3% of Marketing Budgets to AI, But Only 30% Are Ready to Scale AI Capabilities, Gartner, 11 May 2026
- Peter Field's seven lessons for advertising during a recession, IPA
- Advertising in a Recession, commscouncil.nz
- Strategies for marketing through a downturn, Thinkbox, 17 July 2023
- Modern Marketing Dilemmas: How should marketers stand up to recession?, kantar.com
Written by the CMO Tribune Desk and edited by Lila Odinaev. Every factual sentence was checked against a passage on the source it cites. See how we work, or report an error on the corrections page.