Why We Consolidated 34 Campaigns Into 6 (and ROAS Jumped)

We inherited an account with 34 active campaigns on Meta. Different manager, different intern, a new "let's test this" idea every quarter, and nobody ever turned anything off. The account was spending about $85k a month and blended ROAS sat at 1.9. That is exactly the situation where campaign consolidation stops being optional and becomes the biggest lever you have. We cut it to 6 campaigns over three weeks, and ROAS climbed to 3.1. Here is what actually happened and why.

Why 34 campaigns was quietly bleeding money

The problem with a sprawling account is not the number itself. It is what the number does to your data. Meta's algorithm optimizes on conversion signal, and it needs a decent volume of that signal per ad set to get out of the learning phase. The rough benchmark is around 50 conversions per ad set per week. Miss it, and the ad set either sits in "learning limited" or re-learns every time you touch it.

Now split a fixed number of monthly purchases across 34 campaigns and however many ad sets sit underneath them. This account had roughly 90 ad sets. The math is brutal. A store doing 1,200 purchases a month has maybe 300 a week. Spread that across 90 ad sets and you average about 3 conversions per ad set per week. Almost nothing was ever properly optimized. Everything was starved.

That is signal dilution, and it is the quiet killer in most bloated accounts. You are not buying worse traffic. You are buying the same traffic with an algorithm that never gets enough feedback to tell good from bad.

The second cost was audience overlap. Half those campaigns targeted overlapping interest stacks and lookalikes, so we were bidding against ourselves in the same auction and paying more for placements we would have won anyway. Consolidation fixes both problems at once, which is why the effect on ROAS tends to be bigger than people expect.

How we decided what to keep

Before touching anything, we pulled 90 days of data and sorted every campaign by spend, purchases, and ROAS. Three buckets fell out fast.

• Real performers with enough volume to stand alone (a handful of prospecting and retargeting campaigns).
• Zombie campaigns spending $200 to $2,000 a month on three or four purchases, with no statistical meaning.
• Genuine duplicates, where two campaigns did the same job under slightly different names.

The zombies and duplicates were about 28 of the 34. That is the pile you fold in. We mapped the survivors to a clean layout: broad prospecting, a couple of interest and lookalike groups worth keeping separate, retargeting, a catalog/DPA campaign, and one testing campaign walled off so experiments never touch the money. The full logic we use for this lives in our guide to Meta ads account structure.

Consolidating without torching the account

The mistake people make is deleting 28 campaigns on a Monday and rebuilding from scratch. You reset every bit of learning at once and spend the next month in freefall while the new setup gathers signal. We stage it instead, and we watch marketing efficiency ratio alongside ROAS so platform-reported numbers do not fool us. There is more on why that gap matters in our breakdown of ROAS vs MER.

Our sequence looked like this:

1. Build the 6 target campaigns and run them alongside the old ones for a few days, so the new ad sets bank conversions before anything gets cut. 2. Turn off the clearest zombies first, the near-zero-volume ones. Nothing to lose there. 3. Move budget from paused campaigns into the survivors in steps. Big same-day budget jumps knock an ad set back into learning, so we raised spend by roughly 20 to 30 percent every couple of days. 4. Fold the duplicates last, once the consolidated versions had proven they could hold performance.

The whole thing took about three weeks. We never had more than a quarter of the budget in motion at once, so even if a consolidated campaign wobbled, the account held. This is the same staged approach we walk clients through in our paid media services, and it is the least dramatic way to make a big structural change.

What the numbers did

By week four the account had 6 campaigns and blended ROAS had moved from 1.9 to 3.1. The mechanics were simple. Ad sets that used to see 3 conversions a week were now seeing 40 to 60, so most exited learning and stayed out. Cost per acquisition dropped about 28 percent, which tracks with the approach in our notes on lowering CPA. Same creative, same offer, same audiences. The only thing that changed was that the algorithm could finally see what was working.

Two honest trade-offs. We lost some reporting granularity. With 34 campaigns you can point at a line item and say "that interest is soft." With 6, you read performance at the ad set level and lean harder on creative testing. Fair price. And consolidation is not permanent. Accounts drift back toward sprawl as people add tests, so we re-audit structure every quarter and fold anything that crept in.

If your account has drifted past a dozen or two campaigns and performance feels stuck, campaign consolidation is usually the first thing worth checking. We will do it for free. Grab a free ad account audit and we will tell you which campaigns are diluting your signal and what a cleaner structure would look like for your spend.

FAQ

How many campaigns should an account actually have? There is no magic number, but most accounts under $200k a month run fine on a handful. The real test is whether each ad set clears roughly 50 conversions a week. If it cannot, you have too many.

Will consolidating reset the learning phase and hurt performance? It can if you do it all at once. Stage the changes, move budget in 20 to 30 percent steps, and let the new campaigns bank conversions before you cut the old ones. The dip is usually small and short.

Does campaign consolidation work the same on TikTok or Google? The principle holds anywhere an algorithm needs conversion volume to optimize, though the thresholds differ. Fewer, better-fed campaigns beat many starved ones on every major platform we buy on.