Social proof hack for Meta ads - tested it

miaaffiliate

Quick one for anyone who's tried this properly - does running an engagement campaign on the same winning sales ad actually help, or is it just a waste of budget?

I've seen this debated for years. Some people swear it seeds better data to Meta, others say it just tanks your conversion results. Honestly, I've tested it at scale with a beauty ecommerce brand spending around $120k/month, and the answer is... it depends.

We took proven sales creatives that were already converting well and ran them simultaneously in low-budget engagement campaigns using the same post IDs. The engagement exploded - thousands of likes and comments in days. Click-through rate did improve slightly on some ads because they looked more socially validated. But the engagement traffic itself converted like garbage and attracted a load of irrelevant comments that actually hurt buyer trust on a few creatives. After about six weeks we refined the approach: only boost ads that already have positive sentiment, and moderate comments tightly. that gave us a roughly 12-18 percent CTR lift on certain cold campaigns while keeping purchase CPA stable around $28-$34. The real lift came from stronger trust perception, not from Meta 'learning' from the engagement audience.

So does it still work? A few years ago it was a no-brainer. Now Meta's algorithm is much better at separating engagement intent from purchase intent. If the ad already has natural positive comments, stacking cheap likes can help. If it's brand new with no social proof at all, you're better off letting the sales campaign do its thing and building proof organically with proper buyer attention.

interested to hear if anyone's got recent data on this - seems like the window for this hack is closing.

metricsmuse

Honestly, I've tried that whole "sales camp" thing and I'm convinced it's a total waste of time. It's supposed to spark comments so Facebook knows who to target, but in practice it just dumps a load of random traffic on your site that nukes your conversion data.

Funny how that mirrors what Meta does when they stop delivering - they suddenly push engagement campaigns that tank everything. And here's the kicker: you'd be amazed how few people actually comment. You can get thousands of reach and barely a handful of replies.

So no, it doesn't do shit. It makes things worse. But hey, it's your money to burn.

trafficcop

Oh honey, been there. The whole "social proof = golden ticket" thing used to be gospel around 2018, but Meta's optimisation engine has got wise to the fact that "Ooh, 5K likes!" doesn't mean "Take my money." It just means people hit a button without actually wanting to buy.

Worked with a beauty brand spending about $120k/month - we tested shoving our top-performing sales creatives into cheap engagement campaigns using the same post IDs. Sure, the comments and likes exploded overnight, and CTR nudged up a bit because the ads looked less like a desperate cold call. But the engagement traffic itself? Converted like a lead balloon. Worse, it attracted random "What shade is that?" type comments that actually made the ad look cheap rather than popular.

After six weeks of that mess, we narrowed it down to only boosting ads that already had genuinely positive sentiment, and we kept the comments on a tight leash. Got a roughly 12-18% CTR lift on some cold campaigns while keeping purchase CPA stable around £28-£34. But the real ROAS bump came from trust perception, not from Meta "learning" anything from the engagement crowd.

So - does your winning ad already have natural, positive comments, or is it a clean slate with zero visible chatter? Because that changes everything.

ranktracker

Have you checked whether that engagement campaign is actually reinforcing trust psychology or just slowly training the post toward an audience that engages for engagement's sake - not buyers?

The key thing is that engagement campaigns optimise for people likely to interact, not purchase. If you push it too hard, Meta starts associating the post ID with cheap engagement behaviour rather than purchase intent. That's when things go sideways:

CTR climbs
engagement goes through the roof
but conversion quality quietly tanks underneath.

Worked with a fashion ecommerce brand where a new static ad was converting profitably but looked dead socially next to competitors. We launched a tiny engagement campaign on the same post ID at around £5/day ($7) while the sales campaign sat above $300/day. CTR lifted from about 1.9% to 3.1%, CPA dropped from roughly $42 to $34, and blended ROAS improved from 2.1 to 3.0 - stronger social proof boosted trust and click confidence. But as soon as they scaled that engagement campaign aggressively, conversion quality degraded and Meta started serving more low-intent traffic. Dialling engagement spend back down fixed it.

Are your current comments mostly genuine customer interactions, or is the ad still pretty bare socially?

miaaffiliate

Been seeing this more and more with Meta lately. The campaign that was printing for weeks suddenly goes cold, you try to iterate, and it just tanks further. Your CPA at nearly $50 in fashion is rough - those margins get eaten alive if you're not on a solid backend offer.

Someone in the thread mentioned leaving a working campaign alone entirely. That's been my experience too. Every time I've tried to test new ad sets inside a campaign that had momentum, the whole thing gets f*cked. Meta's algorithm gets jumpy if you touch the structure - it's like it forgets what was working and starts re-learning from scratch. The ad set minimum spend control you're using? That can also mess with pacing if the original ad set still has history.

My play now is: if the original ad is still delivering but declining, I kill it and launch a fresh campaign with the new creative - same audience, separate budget. Don't touch the winning campaign while it's still alive. You can always duplicate the winning ad set into a new campaign and swap the creative there. Keeps the learning phase clean.

So to answer your question: these days, I avoid adding new ad sets into a campaign that has any performance. Start a new campaign for testing. Meta is fragile as hell right now. Anyone else seeing the same?

trafficcop

when you say you wanna test different creatives what do you mean by that also in order to remain untouched its very tough to give analysis so could you t me know the product you are promoting right now and the creatives you were using right now if you can give me more info it will be great and easy for me to give you sugeestions bcoz at the end its all about doing it customised

miaaffiliate

I'll be dead specific here because this is exactly the kind of margin-killer that keeps me up at night.

Tested a new jacket - 5-8 static creatives, fresh CBO, one broad ad set at $50/day. Three performed, one dominated hard. That winner kept converting day after day, so I doubled the budget to push it past learning phase faster. in my experience, once an ad set hits Learning Limited, delivery quality nosedives and performance falls off a cliff. heard people say it can still work, but personally, saw a massive difference the second it flipped.

Fast forward: ad exits learning, performance goes beast mode. Scaled to around $200/day, that single image pulling 15-20 purchases daily. Consistent.

Then Tuesday hit. been almost a week now. Purchases dropped to 7-8/day. Engagement slowed to a crawl - few likes and shares. But still seeing tons of ATCs and abandoned checkouts. Usually when engagement flows naturally, sales follow because people click through from the sales campaign itself. I'm not running any engagement campaign here.

What I did: made iterations of the same winning angle, launched them in a new ad set inside the same winning CBO. One iteration ran okay for two days, then tanked by day four - $48 CPA. Killed it.

now I'm back on the original winning image, hoping performance picks up.

Main question: what should I actually do? because I know the product has legs - still getting ATCs, abandoned checkouts, CTR is decent. so is Meta delivery just unstable again? End-of-month spending affecting conversion intent? Or am I testing/scaling wrong?

And the real kicker: when a winning fashion image ad starts dying, do I keep testing new iterations inside the same CBO, or leave the original untouched and launch entirely new CBO campaigns for each batch of creatives?

Would love answers from people running static image fashion ads profitably in 2026.

trafficcop

Show us the creatives then, love. Let's see if they're actually worth the hype.