When to run this checklist
Run it when marketing spend has to fall during a period, after the size of the cut is known and before any budget, target or campaign is changed in the accounts. At that point every line is still cheap to settle: the ranking can be redone, a commitment can be found before it is broken, and the starting settings can be recorded while they are still the starting settings. Once the first changes are made, the evidence for the rest is already noisier, and a cut made in one afternoon cannot be read or reversed in order.
The person who ticks it is whoever owns the marketing budget, usually the marketing lead, with finance for the aim, the commitments and the restore triggers, and each channel owner for the account lines. Three guides own the method and agree on the same layer of checks: the cross-channel cutting guide for where a cut should land across channels, costs and teams, cutting a Google Ads budget for the order of changes inside Google Ads, and reducing spend on Meta for lowering, consolidating and pausing on Meta. This list compresses what they share; for the reasoning behind any line, follow its guide.
Neighbouring pages sit after the decision rather than before it. The monthly paid media review checklist agrees next month's split once results are closed, and the budget month-end close checklist reconciles what was spent with finance. Rebuilding the year's numbers after a cut is the reforecasting tutorial. If the cut also changes what a platform optimises towards, the conversion goal change checklist gates that change separately.
Before you start
- The size of the cut and its period, as finance has stated them, and what the money is needed for.
- The approved budget and the tracker, ideally a budget tracker with a commitments log, showing committed, spent and invoiced money for every line.
- Every contract and order with its value, its notice terms and its next break point.
- Each channel's history of budget changes, with the results that followed, and any incrementality test results.
- The business's own numbers: what a customer or a qualified lead is worth, and the contribution by product, customer group and channel.
- The phone, offline and CRM records that show results the ad platforms do not.
- The list of paid tools and seats, with renewal dates.
- A copy of the budget reallocation memo ready to fill in.
The checklist
The groups follow the order of the decision: agree the aim, gather the evidence at the margin, list what cannot or should not be cut blindly, plan the order of changes, then set expectations and the route back. A line in the first three groups that cannot be ticked stops the cut until it can; the last two are settled before the first setting is changed, not after.
The aim of the cut
Evidence at the margin
Commitments and protected work
Sequencing in the accounts
Expectations and the way back
Where each check is explained
The widget keeps the lines short so they can be ticked; the reasoning lives in the guides. Each check links to the page that explains it.
| Check | Explained in |
|---|---|
| Aim agreed with finance | Cutting a marketing budget across channels |
| Surviving objectives listed | Cutting a marketing budget across channels |
| Sources of contribution marked | Cutting a marketing budget across channels |
| Lines ranked by marginal return | Cutting a marketing budget across channels |
| Marginal spend found inside each platform | Reducing spend on Meta |
| Brand search decided on evidence | Cutting a Google Ads budget |
| Phone and offline sales checked | Cutting a Google Ads budget |
| Commitments listed with break points | Cutting a marketing budget across channels |
| Headroom measured against commitments | Tracking marketing spend against budget |
| Slow-to-rebuild cuts decided in writing | Cutting a marketing budget across channels |
| Measurement, exclusions and creative kept | Reducing spend on Meta |
| Unused seats and duplicate tools found | Reviewing marketing tool licences and seats |
| Exclusions first | Cutting a Google Ads budget |
| Trim rather than pause what pays on average | Cutting a Google Ads budget |
| Cut inside a portfolio first | Cutting a Google Ads budget |
| One target change at a time, apart from budget changes | Cutting a Google Ads budget |
| Overlapping ad sets consolidated first | Reducing spend on Meta |
| Fewer tests, run properly | Cutting a marketing budget across channels |
| Changes staggered and noted | Cutting a marketing budget across channels |
| Lost volume forecast from marginal cost | Cutting a Google Ads budget |
| Delayed effects named | Cutting a marketing budget across channels |
| Memo written and approved | Budget reallocation memo template |
| Restore triggers agreed in advance | Cutting a marketing budget across channels |
| Starting settings recorded | Cutting a Google Ads budget |
| Each step judged on the business's numbers | Reducing spend on Meta |
If an item fails
The first group is the hard stop. If finance and marketing have not agreed whether the cut is for cash this period, profit for the year or a lower cost base, the ranking that follows answers the wrong question: a cash cut cares when money leaves the business, so commitments dominate, while a profit cut cares what each line would have returned. Do not let channel owners propose cuts until the aim and the surviving objectives are written down, or the cut falls on whoever argues least.
In the second and third groups, a missing commitment list or a missing ranking also stops the cut. A cut measured against invoices alone looks bigger than the money actually free, because much of what remains may already be promised; the guide to tracking spend against budget explains why headroom is the budget minus Committed SpendMoney already promised to suppliers through a signed order, booking or statement of work, tracked apart from what has been spent or invoiced.. Ranking by average return rather than at the margin is the commonest failure: the marginal returns and forecasting guide explains why the last increment almost always earns less than the average, and the ROAS calculator turns each slice's spend and revenue into a Marginal ROASExtra revenue returned by the last increment of ad spend. to compare with what a customer is worth. The ROAS calculator page for budget cuts works through ranking each line's last slice against break-even and forecasting the revenue a cut gives up. Some lines can go ahead with a note. Brand search and retargeting rarely have an IncrementalityConversions caused by marketing. test behind them; keep them at a level the memo records as an assumption, rather than cutting or protecting them on the reported return. A cut to brand reach, the email list or search content can be made, but only as a written decision with its delayed cost: the cross-channel guide reports that the IPA's published analyses have argued that brand-building effects build over long periods and persist.
In the sequencing group, a plan that changes everything on one day goes back for re-ordering. The guides record what the platforms document: Google's Smart Bidding pages list large changes to budgets and targets among those that can start a learning period, and Meta's learning phase documentation lists large budget changes and extended pauses among the edits that can restart the Learning PhaseThe period after launch or a significant edit when an ad set's delivery is still being optimised., without either publishing where the line falls. How a lowered campaign budget is spread across ad sets on Meta is set out in the Meta guide from Meta's documentation on Advantage+ campaign budget. The step sizes and the waits between steps are a judgement for each account, not a rule this list can give. A test that cannot reach a readable answer at the lower budget is paused rather than run thinner, as the 70-20-10 budget rule advises.
The last group must be finished before the first change. A cut with no restore trigger tends to become the new budget; the scenario planning framework gives the structure of a signpost, a threshold and an agreed response. A forecast built from average costs overstates what a targeted cut will lose and leaves leadership judging the result against the wrong number; redo it from the marginal cost per result, and let the CPA calculator do the arithmetic. And if the starting settings have not been recorded, record them now: once the first change is made, they are no longer the starting settings. When a trigger is later met, the tutorial on restoring paid media spend after a cut uses that record to bring the paid lines back in order.
A kitchen installer cuts a quarter's spend (illustrative)
A £9,000 cut that skipped the equal share
Gwembrax Kitchens, an invented fitted-kitchen installer with a showroom, is told by finance to take £9,000 out of a £45,000 quarterly marketing budget, for cash this quarter. The first draft cuts every line by a fifth. The list stops it at the ranking. Last quarter's budget steps show the last increment of category search costing about £410 per booked design appointment against an average of £260, and Meta prospecting about £520 against £300; the business can afford about £400 for an appointment.
| Line | Quarter budget | Decision | New budget |
|---|---|---|---|
| Category search | £15,000 | Exclusions first, then one target tightened | £11,500 |
| Brand search | £2,000 | Kept; untested, recorded as an assumption | £2,000 |
| Meta prospecting | £9,000 | Two overlapping ad sets consolidated, then lowered | £5,500 |
| Showroom remarketing | £2,500 | Kept; a spring holdout showed it added appointments | £2,500 |
| Email programme | £1,500 | Kept | £1,500 |
| Home-interiors magazine series | £6,000 | Signed; break point falls after the quarter | £6,000 |
| Content and photography | £3,000 | Kept | £3,000 |
| Tests | £4,000 | One of two tests paused | £2,500 |
| Tools | £2,000 | Unused seats cancelled at renewal | £1,500 |
| Total | £45,000 | £36,000 |
The commitments line changed the plan most: the magazine series could not move, so the equal-share draft would have promised £1,200 of savings that did not exist. A local-area search campaign looked expensive in the platform, but the call log showed it booking appointments by phone, so it was left alone. The changes ran over six weeks: exclusions in week one, the Meta consolidation in week two, the search target in week four, each noted with its reason. The memo forecast about 15 fewer appointments from the marginal costs, not the 25 the averages implied, and finance agreed in advance that £3,000 returns to category search if next quarter's confirmed orders pass 40 by the end of its second month.
Common mistakes
- Cutting every line by the same share. Good and poor spend go together, and commitments that cannot fall are counted as savings anyway.
- Ranking on average return. Saturated lines look safe and lines with room to grow look expendable, against Diminishing ReturnsEach extra unit of spend or effort buys less extra result than the one before it..
- Measuring the cut against invoices. The money already committed is not free to cut, whatever the invoices say.
- Letting the ranking decide brand reach by default. Work that is slow to rebuild needs a written decision, not a side effect of a short-term report.
- Cutting measurement and exclusions to save money. They cost little and make every remaining pound readable.
- Making every change on the same day. Nobody can say which cut cost what, and learning periods stack.
- Ignoring where the contribution comes from. A cut that lands on the products and customers that earn most of the Contribution MarginRevenue left after every cost that varies with each sale. costs more than its size suggests.
- No starting settings and no restore trigger. The cut becomes the budget, and nobody can undo it in order.
The checklists library holds the lists that sit beside this one, from the monthly paid media review to the month-end close.
