Every active campaign is quietly asking the same question: more budget, or less? Most of the time this gets decided by feel, a campaign "seems to be doing well" so it gets more spend, or "seems to have stalled" so it gets cut. Feel isn't nothing, but it's an unreliable way to make a decision that directly moves money, especially because the signals that should drive a pause decision and the signals that should drive a scale decision are genuinely different, not opposite ends of the same scale.
When to Pause (or Seriously Reconsider) a Campaign
CPA has been consistently above your breakeven for a sustained period, not just a bad day or two. One rough day is noise. A week or more of CPA meaningfully above what you actually need to stay profitable, with no clear temporary cause, is a real signal.
Creative has shown the fatigue signature and hasn't been refreshed. Frequency climbing, CTR sliding, CPM flat, if that pattern has been present for a while and nothing has changed in response, more time isn't going to fix it. New creative might; more patience won't.
The campaign has never stabilized out of learning phase. If it's been repeatedly reset by edits and has never had a clean run, the "bad performance" you're seeing may not reflect the campaign's real potential at all, it may just reflect instability. Worth fixing the instability (stop editing it) before deciding the campaign itself is the problem.
It's competing with another campaign for the same audience. If two active campaigns are targeting overlapping people, pausing or consolidating the weaker one often improves the other's performance too, since they stop bidding against each other.
When to Scale a Campaign
CPA has been comfortably below breakeven for a sustained period, with stable delivery. Consistency matters here, a campaign that's been reliably profitable for a couple of weeks, not just a single good day, is a much safer scaling candidate.
Frequency is still low to moderate, with room before fatigue sets in. Scaling a campaign that's already showing early fatigue signs just accelerates the audience burning out, better to refresh creative first, then scale.
The audience isn't yet showing saturation signs. If cost-per-result has been flat or improving even as spend has grown, there's likely still room in the current audience. If cost-per-result has been climbing as spend grows, that's an early saturation signal worth addressing before adding more budget on top of it.
The Actual Decision Between Two Scaling Methods
If a campaign qualifies for scaling, the next question is how, increase the existing budget, or duplicate the ad set as a fresh one.
Increasing budget works better when the campaign is stable and you want continuity, keeping the same learning-phase progress and creative performance history intact. A rule of thumb some media buyers use is limiting budget increases to roughly 20% at a time to avoid knocking the campaign back into a fresh learning phase, though Meta's own systems have gotten somewhat more tolerant of larger changes over time, treat this as a starting guideline, not a hard rule, and watch delivery stability after any change you make.
Duplicating the ad set works better when you want a genuinely fresh test, a new audience variant, a different creative angle, or simply isolating new budget from an existing campaign's learning-phase history so a bad interaction doesn't disrupt what's already working. It's slower to ramp (a new ad set starts its own learning phase from zero) but lower-risk to the original.
Neither is universally correct; the choice depends on whether you're trying to extend something that's working cleanly, or test whether it works at a meaningfully different scale or angle.
Why This Decision Deserves More Than a Gut Call
Pausing a campaign that was actually about to recover from a temporary dip wastes the learning-phase progress and creative-testing history you'd already built up. Scaling a campaign that was about to fatigue just accelerates wasted spend. Both mistakes are common, and both come from treating "is this working" as a single, current-moment judgment instead of a trend-based one.
The framework above isn't about eliminating judgment, it's about grounding the judgment in specific, checkable signals (CPA trend against your real breakeven, the fatigue signature, saturation trend, delivery stability) instead of a general sense of whether a campaign "feels" like it's working.
Where This Gets Faster
Pulling CPA trends, the fatigue signature, and saturation signals for every active campaign, every time you're deciding whether to pause or scale something, is real, recurring work, and it's exactly the kind of pattern-check that doesn't need to be done manually every time.
This is what WizeScale's recommendation engine is built to do: it reads your account's actual signals, CPA trend against a computed breakeven, fatigue indicators, saturation patterns, and surfaces a specific, ranked recommendation (pause this, scale that, watch this one) instead of leaving you to run the checklist above from scratch on every campaign, every week.
Pause or Scale, Backed by a Reason
Pausing and scaling aren't opposite ends of one "is it working" judgment, they're driven by different, specific signals. A campaign worth pausing usually shows a sustained CPA problem, clear creative fatigue, or instability that's never let it prove itself. A campaign worth scaling shows sustained, stable profitability with room left before fatigue or saturation. Check for those signals specifically before deciding either way.
For a ranked pause/scale recommendation instead of running this checklist yourself, see your Meta Ads recommendations with WizeScale. Read-only, built on the same signals described above.
