Facebook Ads Digest, June 2026: Ban Storms, the Death of Copy-Paste Creatives, and Why Tier-3 Saves ROI

I'll be honest — this June was a rollercoaster. The month started almost calm; we'd even caught our breath after a rough May. Then, somewhere after the 10th, real hell broke loose. The good news: we got through it with solid numbers. The bad news: the rules of the game have changed for good, and I want to share what we saw from inside the team.

1. First, about the storm

June 10th — a wave of ad-account bans. The 12th — Meta just lay down to rest. The 13th — social profiles and logouts started raining down; selfie checks began mowing down social accounts in batches, and on sharing, Business Managers flew off. And it's not just me seeing this. I compared notes with several teams, and everyone describes the exact same picture, day for day. Some guess it's because of the World Cup — auction competition right now is insane. Maybe. But the cause matters less right now than the response.

And everyone's response is the same, and it's simple. A full overhaul of the launch structure: fresh fan pages, new pixels, new domains, lots of cabinets in rotation. Plus aggressively killing whatever doesn't catch, and fresh launches every day instead of hoping for one eternal combo. Rejects have become background noise you just get used to. Submit an appeal, the ad goes live, spends a little, gets rejected again — round and round. It's the new routine, and the sooner you accept it as normal, the calmer you work.

2. The discovery of the month: copy-paste is dead

This is the one I want you to remember best. It used to be simple: find a live creative in a spy tool, resize it a bit, swap the text or the button, and run it calmly. Now that approach works worse and worse — and it's not just about competition.

Meta has completely rebuilt its delivery system. Facebook and Instagram no longer work as a targeting platform where audience interests were king. It's now an AI-based recommendation system, very similar in logic to TikTok. The algorithm decides in real time which creative even makes it onto the short list shown to a specific person. And that's where the problem starts for reworked creatives.

The system has learned to read a creative deeply: visual structure, hooks, objects in frame, the first seconds of a video, storytelling. To the algorithm, two creatives can be practically identical — even with different text, voiceover, or button — if they keep the same visual pattern. So it happens: a spy creative looks great in the library, it works for someone, you launch the same thing and get no adequate results. Just expensive. Meta has already accumulated signals on that pattern and quickly flags burnout.

I summed up the takeaway for the team like this: the winner isn't whoever found the creative in a spy tool first, but whoever can quickly make new angles, new storytelling, new hooks. We still need spy tools — but as an analysis instrument, not a warehouse of ready-made combos. They show which offers are alive, which emotions work, which video structures land. And then we produce that ourselves.

3. A small but very interesting detail about design

I dug into old folders of creatives I made when I was just getting to know Canva. Objectively they looked worse — weaker composition, imperfect typography. I retested a few and made similar variations. And you know what? Those "clumsy" creatives performed far better than their polished AI versions. In some cases the difference in cost per install reached five times.

I'm not saying you should now deliberately make bad creatives. My point is different. After hundreds of hours in spy services, a buyer unconsciously starts copying the market; their work begins to look like everything else in the feed. A newcomer looks at the task through the eyes of an ordinary user, and that's exactly why some "wrong" decisions break banner blindness better than yet another perfect creative. Sometimes the most expensive skill isn't doing it better than everyone — it's keeping a fresh perspective.

4. On setups and numbers

Here, like the whole market, we converged on the classics. One CBO, one broad ad set, no unnecessary audience exclusions, and ten to twenty creatives inside. If you split into five ad sets of five creatives each, nothing stabilizes. Scaling is conservative — no more than +20% of budget per day, otherwise the campaign flies back into learning. We duplicate winners rather than scaling them up sharply. A broad audience plus AI optimization simply works better than the old manual schemes right now.

On the money, the picture is this. On a working Tier-1 GEO, the cost per first deposit held around two hundred dollars — and that's with stable conversion, players playing. Reg2Dep across different flows ranged from seventeen to fifty percent, with a median around a third. Unit economics jump day to day: one day you've got campaigns at nearly 200% ROI, the next a night test goes -40%. That's normal for a storm — the key is to look at the week, not a single day.

And one more important thing. Tier-1 is unstable right now, so we took Tier-3 into rotation. At the start there were questions about the approaches, but now I can say June is going confidently there, and the cost per conversion is several times lower — from ten to twenty dollars. We run for quality, with a decent average check, no celebs and no crash, because the job is to give the advertiser quality traffic, not to squeeze a short-term number at any cost.

5. Check your pixel, seriously

A separate story that sounds like mysticism, but I have to tell it. One of our buyers struggled for weeks with expensive CPM and CPC. He changed everything: fan pages, domains, creatives, accounts, Business Managers. Nothing helped. The only thing that stayed the same was the old pixel he'd been running for months. He created a fresh pixel on a fresh BM, launched the same creative — and CPM returned to normal while CPC became several times cheaper.

I wouldn't have believed it myself, and the industry is still arguing whether it's the pixel or the cabinet. There's no consensus. But if your CPM has been expensive for more than a week and nothing fixes it, test a fresh pixel. It's a cheap experiment with a potentially huge payoff.

6. What I'm taking away from June

The market has finally shifted from copy-pasting combos to production. Advertisers are tightening KPIs, margins are under pressure, ROI isn't the fat number it once was. But that's no reason to whine — it's a reason to work differently. Produce your own approaches, keep a fresh perspective, diversify GEOs, don't be afraid of Tier-3, and treat traffic quality the way an operator sees it — through repeat deposits and real profitability, not just cost per lead.

June was hard. But months like this are exactly what separate those who just copy from those who truly understand why a creative worked. And I'm glad we're in the second group.