How should we reallocate budget mid-month when results swing?
Our answer
Check spend and orders each day to catch breakage, but move budget weekly, not daily. Shift money only on 7-day results that held all week, in steps of about 10 to 20%, toward what your tests favor.
- Confidence
- Medium
- Strongest proof
- 6 official rules
- 1 dataset, 7 practitioners, 11 opinions or claims
- Sources
- 19 sources
- 25 statements
- M
G
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentDaily numbers swing by design. Meta may spend up to 75% over a daily budget on some days and suggests reading a full week (Meta daily budget overspend). Google says to judge a change over 2 conversion cycles or 2 weeks.
Short-term dips are often randomness, site issues or event delays, and forcing budget elsewhere usually does more harm than good (Jon Loomer, Meta ads educator). Flighted (paid media agency) checks pacing daily but scales weekly.
Big moves have a cost. Meta says a large budget change may restart learning, and PPC Hero sees cost per purchase move 15 to 30% for days after edits. Sellforte (marketing mix modeling firm) advises 10 to 20% changes per channel.
Our judgment
- Check daily, but move budget weekly. Read spend, orders and tracking every day, then move money on rolling 7-day results judged against store revenue and new-customer cost, because daily results swing by design, short-term dips are often noise or breakage, and store totals are what the month must match.
- Move in steps of about 10 to 20% per channel. Treat holding as a decision too.
- Move money between channels on what tests say. Go by each channel's next dollar, not one platform's reported swing.
- To land the month on plan, trim or add daily budgets against weekly spend targets rather than pausing and restarting.
- Daily budget scaling doesn't change the advice to move budget weekly. Nick Theriot's daily 20% raises and True Classic's daily agent moves aren't compared with weekly moves, and Meta's own page describes daily swings that even out over a week.
Our judgment, built from Meta, Google, Jon Loomer, Flighted, Trackbee, Sellforte, PPC Hero, Stella, Grüns, Elevarus, Top Growth Marketing, Nick Theriot, DTC Newsletter / DTC Podcast and Recast: our reading of these sources, not a finding from any one of them.
What the sources say
25 statements from 19 sourcesMeta says a daily budget is averaged over a Sunday to Saturday week: it may spend up to 75% over on a given day, but no more than 175% in a day or 7 times the daily budget in a week.
“For every day ending at midnight, spending won’t be more than 175% of your daily budget.”
Northbeam, an attribution seller that sponsors the show, says its median brand gets 42% of Black Friday revenue from earlier touchpoints, and warns that a strong midday read can tempt brands into overspending.
“median brand pulls 42% of Black Friday revenue from earlier touchpoints”
Google says budget or target changes take effect at the next auction without triggering learning, but advises against several target changes in one conversion cycle and judging over 2 cycles or 2 weeks.
“Avoid making multiple target changes within a single conversion cycle unless there is a clear business need.”
Hook Agency says HVAC service calls drop about 30% in the spring and early fall shoulder seasons, when cost per lead runs 20 to 40% below peak, per its own data with no sample given.
“If demand is going down 30% in January, reduce spend by approximately 30%.”
Nick Theriot raises daily budget by 20% every day that the previous day came in below target CPA.
“all we do is literally increase daily budget by 20%, and then we'll just literally do that every day that yesterday was below that CPA.”
Meta says it may spend up to 75% over a daily budget on some days but no more than 7 times the daily budget in a week, and advises reading results at the campaign level over a full week.
“On a weekly basis, we won't spend more than 7 times your daily budget.”
Meta says targeting, creative or optimization event changes and adding a new ad are always significant edits, while a budget change may or may not restart learning, depending on its size. No safe percentage is given.
“if you increase your budget from $100 to $101, that isn't likely to cause one or more ad sets to reenter the learning phase.”
Meta says ad sets in the learning phase are less stable and usually cost more per result, so wait to edit until learning ends and avoid frequent budget changes, but do not try to avoid learning entirely.
“During the learning phase, ad sets are less stable and usually have a higher CPA.”
Google's Tal Akabas says changing the budget or target on a steady campaign does not start a learning period; only new campaigns learn, so launch them a few days before a promotion.
“You can change your bids or your targets confidently on an existing campaign and know that nothing's going to reset in the backend”
Jon Loomer says a budget increase of around 15% is a safe step that should keep volatility low when scaling.
“A safe increase that will hopefully result in minimal volatility will be around 15%.”
Gruns' growth head matches the tool to the decision: incrementality tests, a mix model and marginal return for big budget calls, and attribution and platform data for daily moves.
“I'm also not letting attribution decide whether Meta deserves another $500K.”
Stella, which sells testing, says a budget call should state direction, size and a check date, with breakeven at 1 divided by margin (a 40% margin means 2.5x). Its step-size examples are illustrative.
“Cutting a channel because the point estimate looked soft is how brands kill channels that were working.”
Sellforte, which sells a mix model optimizer, suggests limiting first changes to 10% to 20% per channel and widening only when recommendations hold steady across model runs.
“Start with 10% to 20% spend changes per channel, then move farther only where the team can execute and the evidence remains credible.”
Luke Austin says marketing forecasts are always wrong, so teams need to keep adjusting to correct course.
“Forecasts are always wrong.”
Michael Kaminsky of Recast says that if forecasts built on last-touch numbers keep missing, those numbers are wrong; your view of marketing performance should match month-end top-line results.
“Your beliefs in terms of marketing performance should be consistent with the top-line results that you see at the end of the month.”
Trackbee suggests watching revenue, orders, MER, spend and CAC daily, the new versus returning split and channel results weekly, and cohorts, LTV and margin monthly.
“If a number would not change what you do next, it belongs in a monthly report, not on your daily screen.”
Haus simulated a year of test-driven budget moves 36 million times: precise tests beat doing nothing 82% of the time vs 62% for noisy ones, and 15 noisy tests a year did no better than 6.
“The precise scenarios beat a frozen-budget baseline 82% of the time. The noisy ones, only 62%.”
Jon Loomer says to judge Meta results on seven-day windows at least two weeks in, not daily swings, since short dips are often noise, and that manual intervention usually does more harm than good.
“Evaluate seven-day windows at least two weeks in rather than reacting to daily swings.”
AdAdvisor, which sells a Meta reporting tool, says to move 20 to 30% of a weak ad set's budget once ROAS sits below break-even for 3+ days, review weekly, and wait 3 to 5 days between changes.
“Move 20-30% of the underperforming ad set's budget to the stronger one. Not all of it.”
True Classic's CEO says it runs 100% of its Meta spend through Triple Whale's autonomous media buyer, Moby 2, on an episode with Triple Whale's COO. No results or price are disclosed.
“True Classic now runs 100 percent of its Meta spend through an autonomous media buyer, Moby 2”
Elevarus, a paid media agency, says not to move budget on Reddit's own higher attribution numbers, since Reddit gains if they are true; test with a geo holdout across matched states instead.
“Treat Reddit Dual Attribution as a hypothesis generator, not a conclusion.”
Top Growth Marketing, which sells media management, says to raise spend only after CPA holds 3+ days, check total spend against weekly targets, and never pause the whole account to restart at month end.
“Only increase spend after CPA holds for 3+ consecutive days”
PPC Hero's Bia Camargo says Google can spend up to 2x a daily budget on one day, and in her accounts budget edits move CPA 15% to 30% for three to seven days. She moves budgets in 20% steps.
“Hold for 5 to 7 days between changes so Smart Bidding can stabilize before you move again.”
Jon Loomer says to judge ads over 7-day windows and not react to 2 to 3 day dips after budget or creative changes.
“you should not be looking at those results for a one day period, two day.”
Flighted, an agency, says to check pacing daily but make scaling calls weekly on blended numbers, give every audience a full week before pausing, and prune account structure monthly.
“Check pacing daily, but make scaling calls weekly on blended numbers.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer