Meta's New Creative Workflow Reports Per Asset. Flexible Format Never Did.

Meta started removing Flexible Format from new ad setups in March 2026 and rolled out a replacement creative workflow that lets advertisers upload up to 10 images and videos per ad, with three aspect ratios per creative and per-asset performance reporting. Existing Flexible ads keep running, but new and duplicated ads use the replacement. Account naming conventions tied to format labels will break, and Q3 historical comparisons go with them.

I've been waiting for someone at Meta to admit Flexible Format wasn't really a format. It was a delivery shortcut bolted onto the old multi-creative ad object, and the reporting Meta gave you for it was almost insulting. The replacement workflow Jon Loomer flagged this week is the thing agencies asked for in 2023. The timing of the swap is the thing they didn't.

What the new workflow actually changes

The Jon Loomer breakdown of the new workflow is the cleanest writeup I've seen so far. The replacement gives you five things inside a single ad object:

• Upload up to 10 images or videos per ad
• Three cropped aspect ratios per creative
• Customizable text and URL destination per creative
• Placement opt-in and opt-out per creative variation
• Performance reporting by creative asset

The headline feature, even if Meta isn't pitching it that way, is the last bullet. Flexible Format pooled performance into a single ad-level number. You couldn't tell which of the variations Meta delivered actually drove the conversion. You could see which got delivered, sure. Which one moved the metric? Black box.

The new workflow breaks that out at the asset level. Which is what creative testing programs have needed since the day Flexible launched. The cost of getting it is that the entire format you built your testing program around isn't selectable anymore.

What broke when Meta swapped formats mid-cycle

The replacement only applies to new and duplicated ads. Existing Flexible ads keep running for now, per Campaign Builder's reporting on the March 2026 removal date. So agencies and in-house teams are sitting on a split account: legacy Flexible ads with one reporting model, new workflow ads with another, and a six-to-twelve-month period where neither side compares cleanly.

The naming conventions that teams built around format labels (FLEX_BROAD_PROSPECT_05, FLEX_RT_INTEREST_22, the whole inventory) point to a format that doesn't exist in the new ad setup. They still parse, but the moment someone duplicates one of those ads into the new workflow, the prefix is a lie. From what I've seen, agencies have two ways out:

Option one. Keep the legacy ads alive as long as Meta lets them run, run the new workflow in parallel, and treat the two reporting models as separate experiments. Pro: no rebuild. Con: every monthly client deck needs a footnote explaining why two ads with the same naming prefix have different reporting fields.

Option two. Rebuild the active set in the new workflow this quarter, accept the data discontinuity, and tag the rebuilt ads with a new prefix that signals which reporting model they belong to. Pro: clean Q3 going forward. Con: you eat the launch and learning phase on every rebuilt ad, which on accounts with thin daily conversion volume is a real cost.

I'd lean toward option two for accounts with enough daily conversions to clear learning quickly. For accounts that already struggle to clear 30 conversions per ad set (which, as Andromeda's 14-day creative lifespan already made painful), option one might buy you a quarter to plan the rebuild around a slower period.

The asset-level reporting Flexible Format never gave you

The under-talked-about part of this swap is that Meta is finally admitting the old breakdown was useless. Flexible let you upload several creatives, trusted the algorithm to pick a winner per impression, and then handed you a delivery report that said which creative got served the most. That's not the same as saying which one converted. Anyone who tried to use Flexible's breakdown to make creative cut decisions ran into the gap fast.

The new workflow ties the conversion event back to the specific creative asset that was served when it happened. So your "kill the underperformers" decision isn't running on delivery share anymore. It's running on actual outcome attribution per asset. This is a non-trivial upgrade, and it's the part of the announcement Meta should be louder about.

The catch: this only works inside the new ad object. The legacy Flexible ads still in your account keep their old, useless breakdown until Meta sunsets them entirely. So the per-asset reporting you've been waiting for arrives gated behind a rebuild.

What the new workflow excludes

Worth knowing before you plan a rebuild. The new workflow does not currently support Advantage+ catalog ads, traditional carousels, or specific conversion-location combinations like Traffic on Facebook/Instagram or Leads with Website plus Instant Forms. Those flows still use their existing setup paths.

So if your account leans heavily on catalog or carousel objects, the rebuild conversation is narrower than "migrate everything." It's "migrate the eligible objectives, leave the rest, and accept that your reporting layer now has three flavors instead of one."

Where this fits in Meta's broader cleanup

This is part of a wider rationalization. Format Display Options started rolling out April 16, 2026, absorbing both Flexible and Collection into a single setup path under the ad creative panel. Meta's framing is that advertisers shouldn't have to think about format selection because the algorithm will pick the right one. The honest framing is that Meta is consolidating format complexity into a smaller surface so it can ship algorithm changes faster without breaking advertiser workflows.

That trade is fine when the consolidation gives you better tools. The new workflow does. It's the migration timeline that's the problem, not the destination.

Three actions worth taking this week

Pull a list of every active ad that uses Flexible Format. Tag the ones that account for more than 10% of monthly spend. Those are the candidates for early rebuild because the per-asset reporting will pay back the learning-phase cost fastest.

Audit your account naming convention for any string that includes FLEX, FLEXIBLE, or FF as a prefix or suffix. Decide now whether you're going to keep using it (and what it means in the new context) or retire it cleanly. Mid-quarter is the worst time to rename a thousand ad sets.

Open one new ad in the new workflow on a low-stakes campaign before your team needs to ship anything important. The setup screen has new fields, new defaults, and at least one place where the previous muscle memory will lead you wrong. Better to find that on a $50/day test than on the launch deck for a major client.

I don't think the agencies that move first here will see a huge edge over the ones that wait two months. But I do think the agencies that wait six months will be doing emergency rebuilds in the middle of holiday season, and that's a worse position to be in than just doing the audit now.

Notice Me Senpai Editorial