Paid social
Your best Meta creative won. Or did Meta just bet on it?
Meta Ads reports results for each image and video in an ad. Why the top row is partly Meta’s forecast, and how to read it before you cut a single asset.
Pierrick NonnisCo-founder, COO4 min read
A Meta ad with several images in it used to say very little about which visual had earned its results. Meta’s newer creative setup changes that. One ad can carry up to ten images and videos, Meta decides which one to show to whom, and Ads Manager breaks the results down asset by asset. For a team that produces creative in volume, it is one of the most useful reports in Ads Manager. After Black Friday, someone will open it and start cutting the visuals that lost. This post is for whoever opens it first.
Where to find the per-asset breakdown in Ads Manager
Jon Loomer has documented the new workflow. In Ads Manager, at ad level, open Breakdown, then Creative, then Media: each image and video gets its own row, with its spend, impressions, clicks and results. Meta has been rolling the workflow out since the spring, for image and video ads (carousels and catalog ads stay as they were), and it applies to new or duplicated ads. If the option is missing, your account or that particular ad may not have it yet.
What it opens up is concrete. You learn which concept carries a multi-asset ad, which format the system keeps reaching for, and whether the expensive shoot earns its place next to the video filmed on a phone. All of that feeds the next brief, which is where creative budgets are won or wasted. Flexible format never offered this. Its creative breakdown, added in 2025, only told single images and videos apart from carousels, and Meta is now replacing the format with this workflow. So the per-asset row is about to turn up in a lot of creative reviews, some of which will run long.
Why the top row is partly a forecast
Each row reports what happened to the money Meta chose to spend on that asset, and Meta does not hand out budget by lottery. Within the first days, the delivery system forms a view of which asset will convert best for which person, and funds that view. The asset on top got the most budget because Meta expected it to do well, which also gives its row the most data on the table and the others the least.
Three things follow from that.
- Each row reaches a different crowd. One asset may have been shown mostly to people already close to buying, another to cold scrollers. Comparing their cost per purchase compares two audiences as much as two visuals.
- Small rows are noise. An asset with €40 of spend and two purchases has a cost per result, technically. It does not have a verdict.
- The assets work as a team. Removing a low-spend asset can change how the whole ad delivers, because the system may have been using it for a slice of people the others do not reach.
In the accounts we run, it is common to see one or two assets take most of an ad’s spend within the first week. That is the system doing its job, and a good reason to read the rest of the table carefully before anyone edits the ad.
How to read it before you cut anything
- Treat spend share as Meta’s forecast. The asset Meta funds is its best guess. A guess that contradicts your creative team’s instinct is worth a test, whichever side you are on.
- Set a floor before judging. Decide upfront how many results an asset needs before its row counts, and leave anything below the floor out of the verdict. We set it per account, at roughly the volume where a week’s cost per result stops jumping around.
- Judge the ad first. If the ad’s overall cost per result is healthy, a quiet asset may be doing useful work. Retire it, watch the ad’s total for a week, and bring it back if the total slips.
- Settle the big questions with a real test. When real money rides on the answer (a new concept against the current one, say, or next year’s production budget), run a Meta A/B test, which splits the audience into random groups that do not overlap. The breakdown gives you the shortlist, and the random split picks the winner from it.
- Keep feeding the ad. The breakdown pays off as a rhythm: every week or two, a new asset in, the starved ones out once they have passed the floor. It is the weekly discipline we described in why UGC wins paid social, with a better dashboard.
The upside goes to the teams that read it well
Meta now shows its work on creative, and a team that reads the breakdown as a forecast can brief faster and put production money where it pays. Deciding what to make next, and what to test before betting a budget on it, takes strategy, execution and a good deal of judgement, and that is where the upside is won or lost. If you want a second pair of eyes on your own breakdown after Black Friday, book a call. We have a soft spot for a table with ten rows.


