If you searched "relevance score facebook ads" expecting to find a 1–10 gauge in Ads Manager, here is the honest answer: that gauge is gone. Meta replaced relevance score with ad relevance diagnostics to make the feedback "clearer and more actionable." This guide covers what replaced it, what each metric means, how relevance drives your costs, and — the part most articles skip — why a high relevance ranking is not the same thing as making money.
What was the Facebook relevance score?
From 2015 to 2019, every ad earned a relevance score between 1 and 10 once it passed a minimum impression threshold. Meta calculated it from "the positive and negative feedback we expect an ad to receive from its target audience." Expected likes, clicks, and shares pushed it up; expected hides and reports pushed it down.
The score appeared after an ad reached 500 impressions, and the core rule was simple: the higher the relevance score, the less it cost to deliver the ad. That relationship still holds today — Meta just stopped expressing it as one tidy number.
What replaced it: the three ad relevance diagnostics
On August 8, 2019, Meta swapped the single score for three diagnostics, each answering a different question about your ad. According to a breakdown of the diagnostics, they are:
- Quality ranking — how your ad's perceived quality compares to ads competing for the same audience.
- Engagement rate ranking — how your ad's expected engagement rate compares to those same competitors.
- Conversion rate ranking — how your ad's expected conversion rate compares to ads with the same optimization goal.
Instead of a number, each diagnostic reports a tier. The same breakdown describes them as "above average" (above the 55th percentile), "average" (the 35th to 55th percentile), and "below average," which itself splits into the lowest 35%, lowest 20%, and lowest 10% of ads. So "below average — lowest 10%" is a much louder alarm than "below average — lowest 35%."
The point of splitting one score into three is triage. A single low number told you something was wrong but not what. The three diagnostics tell you where to look, which is the whole idea behind improving your Facebook quality score rather than guessing.
How to read the three together
Read them like a funnel:
- Low quality ranking points at the creative and the audience's reaction to it — think hides, negative comments, or clickbait patterns.
- Low engagement ranking means the ad isn't earning the clicks, reactions, and shares Meta expected relative to rivals.
- Low conversion ranking points past the click — to the offer, the price, or the landing page — because people click but don't convert.
If quality is fine but conversion ranking is low, don't touch the creative — fix the page and the offer. If all three are low at once, the concept itself is weak.
Why relevance affects your cost
Relevance matters because of how the Meta auction works. For every impression, Meta ranks advertisers by total value, roughly: your bid, multiplied by its estimate that this user takes your action, plus ad-quality signals. A relevant, high-engagement ad can beat a higher bidder and pay a lower CPM, because Meta is optimizing total value — not raw bid.
How much can relevance move cost? An AdEspresso analysis of 104,256 ads, reported by Instapage, found the same ad cost $0.14 per website click at a relevance score of 2.9, and $0.03 per click at a score of 8 — and the higher-relevance version generated four times the clicks on an equivalent budget. That is an old case study on the retired metric, so treat the exact figures as illustration, not a promise. The direction, though, is exactly what the auction math predicts: more relevance, less cost.
When your CPMs climb, relevance is only one of two possible causes. Either the market got more expensive — more advertisers in the auction during Q4 or a sale event, which is nobody's fault — or your ad quality decayed. Diagnosing the difference is the first move in any honest scaling process, which is why it sits at the center of profitable ad scaling.
How to improve your ad relevance diagnostics
There's no button for "raise relevance." You improve the inputs and let the diagnostics follow.
- Lead with a stronger creative. In Meta's current system the creative is the primary targeting signal — the hook, format, and on-screen talent decide who sees the ad more than manual interests do. A scroll-stopping opening lifts engagement ranking first.
- Tighten message-to-audience match. Relevance is relative to the audience Meta shows the ad to. The same creative can rank "above average" for one segment and "below average" for another.
- Reduce negative feedback. Hides and "why am I seeing this" reports are what dragged the old score down and still hurt quality ranking. Avoid clickbait and over-promising.
- Fix the post-click experience. A slow or mismatched landing page tanks conversion ranking even when the ad is great.
- Refresh before fatigue sets in. As frequency rises on a tired creative, engagement falls and relevance slides with it. Watching for Facebook ad fatigue keeps your diagnostics from decaying, and steadily working to increase customer engagement is the same work that lifts engagement ranking.
The trap: relevance is not profit
Here's what every "relevance score" article leaves out. Relevance lowers your cost per click and cost per result. It says nothing about whether the result made you money.
Say you sell a $50 product. Your COGS, shipping, and payment fees eat $25, leaving a 50% contribution margin. Your break-even ROAS is simply 1 ÷ 0.50 = 2.0x. Below 2.0x return on ad spend you lose money on every order, no matter how green your relevance diagnostics look.
Now push relevance up and cut your cost per click in half. Wonderful — but if you scale that winning ad hard, each new dollar reaches a less-responsive slice of the audience. A campaign averaging 4.0x ROAS can hide a marginal ROAS of 0.6x on the last chunk of budget: you added $2,000 of spend and got back $1,200 of new revenue. 1,200 ÷ 2,000 = 0.6, which is well under your 2.0x break-even. Those last dollars lost money while the headline number stayed comfortably green.
Relevance and profit answer different questions. Relevance asks "is this ad cheap to serve?" Profit asks "does the next dollar clear my margin?" You need both, and you scale on the second one. Running the actual numbers with a break-even ROAS calculator turns "my relevance is great" into "I can spend up to here and still profit."
Where PodVector fits
To know your real break-even, you need true per-order profit — revenue minus COGS, shipping, fees, and ad spend — not the ROAS your ad platform reports. PodVector connects Shopify, Meta Ads, Google Ads, Printify, and Printful and computes that true per-order profit across them.
On top of that live data sits Victor, an AI operator that analyzes your numbers and acts on them — with your approval — on the Shopify side. Victor reads your Meta ad data and can tell you which campaigns are profitable at the margin versus which just look relevant, but Victor does not touch your ad account. He's not a dashboard; he's an operator you can put your real profit picture in front of. You can try it here.
FAQs
Does Facebook still show a relevance score?
No. Meta retired the single 1–10 relevance score in 2019 and replaced it with three ad relevance diagnostics — quality ranking, engagement rate ranking, and conversion rate ranking — which you'll find in Ads Manager by adding those columns.
What is a good ad relevance diagnostic result?
Aim for "above average" on all three, which the diagnostics breakdown places above the 55th percentile. "Average" is fine; "below average — lowest 10%" is the urgent one. Because the rankings are relative to competitors, "good" shifts with your audience and industry.
How many impressions before I see relevance data?
The old relevance score appeared after an ad reached 500 impressions, and the diagnostics need a similar volume of delivery before Meta can score them reliably. Below that, don't read too much into an early ranking.
Does a high relevance ranking guarantee lower costs?
It pushes costs down, but it can't override the market. Even a top-ranked ad pays more when auction density spikes — Q4, competitor entry, a big sale event. Relevance is one lever on CPM; auction competition is the other, and only one of them is in your control.
Is relevance the same as ROAS or profit?
No. Relevance lowers your cost per result; it says nothing about margin. An ad can rank "above average" on all three diagnostics and still lose money if your contribution margin is thin. Break-even ROAS = 1 ÷ contribution margin, and you scale on marginal ROAS, not on how relevant the ad looks.