The Budget Reallocation Playbook
Cut, hold, scale rules using verified data.
After we fixed tracking I still made bad calls for a month, because I had good numbers and no rules. I would see a 3.4 verified ROAS and double the budget on a Thursday, then panic on Saturday. The playbook below is what I wrote down so that Thursday-me could not do that anymore.
The one-sentence version
Sort campaigns by verified new-customer ROAS, move 20% of budget from the bottom to the top once a week, and never move money on fewer than 30 verified orders.
The setup
Six campaigns, one Monday
Try it: six campaigns, one Monday
Click a campaign. Get the call.
| Campaign | Spend | Ads Mgr ROAS | Verified ROAS | New cust. | Orders |
|---|---|---|---|---|---|
| Meta prospecting: broad | $12,000 | 1.9x | 3.4x | 91% | 140 |
| Meta retargeting: 30d | $6,000 | 8.1x | 2.2x | 14% | 88 |
| Google brand search | $3,000 | 12.4x | 3.1x | 22% | 120 |
| Google non-brand search | $5,000 | 2.1x | 2.9x | 83% | 60 |
| TikTok prospecting | $4,000 | 0.9x | 2.4x | 95% | 34 |
| Performance Max | $7,000 | 5.6x | 2x | 31% | 95 |
Meta prospecting: broad: Scale +20%
Verified ROAS 3.4x with 91% new customers. Ads Manager under-reports it at 1.9x. Add 20%, check again next week.
Target: 2.5x verified new-customer ROAS. Numbers illustrative.
Good data does not stop you guessing. Rules are what stop you re-guessing.
The rules
Five lines, in order
- Fewer than 30 verified orders? Hold. Read the direction, not the decimal.
- Under 30% new customers? Cap. It is retention. Budget it like a coupon.
- Verified ROAS 20% above target? Scale by 20%. Re-check in seven days.
- Within 10% of target? Hold. Refresh creative before adding money.
- Below target with enough orders? Cut 20%. Watch MER for a week.
Why it matters
Ads Manager ranks them backwards
Look at the table again. By Ads Manager ROAS the ranking is brand search, retargeting, PMax, non-brand, prospecting, TikTok. By verified new-customer ROAS it is almost exactly reversed. The playbook is boring on purpose, because the interesting part already happened when the numbers were verified.
I still get the Thursday urge. The rule about 30 orders is there for me, not for you.
The TrueROAS version
The table is the dashboard
Verified ROAS, new-customer share and verified order count are columns on the campaign table in TrueROAS, next to what the platform reported. Sort by new-customer ROAS on Monday, apply the five rules, close the laptop.
Honest footnote
When the rules break
Launches, sales and seasonality all break the 20% rule on purpose. Suspend the playbook for the event, restart it a week after, and compare against the same week last year rather than last week.
Questions people actually ask
Straight answers
Why 20% steps?
Large budget changes reset Meta's learning and destroy your baseline. 20% is enough to see an effect within a week and small enough to reverse without damage.
What target ROAS should I use?
Your break-even on first-order margin, adjusted for how much lifetime value you are willing to invest. The ROAS calculator on this site works it out from your margin and repeat rate.
How often should I run this?
Weekly, same day, ten minutes. Daily changes chase noise. Monthly changes waste weeks on a campaign that has already stopped working.
What about creative?
The playbook moves money between campaigns. Inside a campaign, the same verified numbers per creative tell you which ads to refresh. Do that first when a campaign lands on Hold.
See your own numbers, verified against your orders.
Free under $10k/month in revenue. 14-day trial above that. Nothing changes in your ad accounts. You will know within a week which platform has been telling stories.
"I built this because I was making $50k-a-month decisions on numbers I knew were wrong. If TrueROAS does not show you something Meta or Google got wrong in the first two weeks, tell me and I will help you cancel."
Rasmus Arvidsson
Founder, TrueROAS. Answers support himself.
Keep going