Free editable template

Kill / Hold / Scale rulebook for Meta ads

Decide once, from your own margin, when an ad gets more money, stays as it is, or gets turned off. Paste an Ads Manager export into the workbook and every row gets a verdict, the rule that fired, and the suggested budget change. The CSV is the rule register to agree with your team.

By GaaS editorial · Sources checked · 15 worksheet rows

Open the CSV in Excel, or import it into Google Sheets and choose comma-separated values. Save your working copy in a controlled location. The CSV contains the rows below and space for your notes; it has no macros, formulas or live account connection. The separate Excel workbook contains editable example inputs and formulas that update the forecast.

How to use the template

Enter your margin once

On the Settings tab, type your average order value, your contribution margin before ads (after product cost, shipping, payment fees and returns) and the share of revenue you want to keep as profit. The tab works out break-even ROAS, target ROAS, break-even CPA and target CPA, plus the minimum spend before judging and the hard-stop spend.

Paste an export

Export ad-level (or ad set-level) results for one date range from Meta Ads Manager: name, amount spent, purchase conversion value, purchases, frequency, link CTR and days live. Paste from row 5 of the Ads tab. Optional columns take each ad's first-week CTR (for the fatigue check) and the daily budget behind the row (for the new budget).

Read the verdict and the reason

Formulas apply the rules in a fixed order: hard stop, data rule, kill floor, fatigue, then the scale test. The Why column names the rule that fired, so you can explain every decision. Colours show SCALE in green, HOLD in amber, KILL in red and WAIT FOR DATA in grey.

Act once, then wait

Turn off KILL rows. Raise SCALE rows by one step and note the date. Add new creative for tired ads and killed concepts. Leave HOLD and WAIT rows alone. The Summary tab shows how much daily budget the kills free up versus what the scales add, so total spend only grows when you decide it should.

Preview the worksheet

Every row shown here is included in the download. Scroll the table horizontally on a small screen.

RuleConditionDefault
Break-even ROAS1 / contribution margin before adsFrom your margin
Target ROAS1 / (margin - profit share you keep)From your margin
Break-even CPAAverage order value x marginFrom your margin
Target CPAAverage order value x (margin - profit share)From your margin
Data ruleSpend at least N x target CPA AND live at least D daysN = 3, D = 3
Hard stopZero purchases after N x break-even CPAN = 3
KillROAS below kill floor after the data ruleFloor = 100% of break-even ROAS
HoldROAS between kill floor and target ROASBand
Hold (promising)ROAS at or above target but fewer purchases than the scale minimum10 purchases
ScaleROAS at or above target AND purchases at or above the scale minimum AND not fatigued10 purchases
Scale stepBudget increase per step+20% (practitioner convention)
Wait between stepsDays before re-judging a scaled ad set3 days
FatigueFrequency at or above F AND CTR down at least X% from launch weekF = 3.5, X = 30%
Budget movesMoney freed by kills vs money asked for by scalesSeparate decisions
Review cadenceSame weekday, same date range, same attribution settingWeekly, last 7 or 14 days

Worked review example

Fictional shop: $60 average order, 40% contribution margin, 10% profit target. Break-even ROAS is 2.50, target ROAS 3.33, target CPA $18, so no ad is judged before $54 of spend and 3 days live. The hard stop is $72 with no purchases.

Of 10 ads, two scale (4.00 and 3.50 ROAS with 52 and 34 purchases), four hold (one tired at 4.00 ROAS, one promising at 3.60 on 9 sales, two in the middle band), two are killed (1.80 ROAS, and $95 with no sales) and two wait for data. Kills free $90 a day; one 20% step on each winner adds $26 a day; the rest funds two new tests.

What this template does and does not establish

The workbook uses platform-reported purchases and revenue, which depend on Meta's attribution setting; it does not prove that an ad caused a sale. The default thresholds are starting points, not Meta rules or industry benchmarks. Built for purchase-optimised campaigns; for lead generation use cost per qualified lead. No account connection, macros or automatic changes: it only suggests.

Primary references

Continue the work