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. According to Meta's original announcement, the score was "calculated based on 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.
According to Segwise, your ad must cross 500 impressions before Facebook calculates your relevance diagnostics. 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 Meta's own help documentation, the three diagnostics 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 competing for the same audience.
Instead of a number, each diagnostic reports a tier. According to Segwise's 2026 breakdown, the tiers are: above average (top 55th percentile and above), average (middle 35th–55th percentile), and below average — which itself splits into the bottom 35%, bottom 20%, and bottom 10% of ads. So "below average — lowest 10%" is a much louder alarm than "below average — lowest 35%."
An important nuance: according to Meta's announcement, these ad relevance diagnostics are not factored into an ad's performance in the auction. They are diagnostic and reporting tools — not direct auction inputs — even though the underlying quality signals that feed them do influence delivery.
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.
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. According to Segwise, Facebook looks at positive signals (watch time, reactions, click-throughs) and negative signals (people hiding your ad, skipping fast, or flagging misleading content).
- Low engagement ranking means the ad isn't earning the clicks, reactions, and shares Meta expected relative to rivals. According to AdSpyder, this is where hook quality and thumb-stopping creative matter most.
- Low conversion ranking points past the click — to the offer, the price, or the landing page — because people click but don't convert. According to AdSpyder, stronger intent match, a simplified landing page, and better tracking all move this number.
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.
Where to find relevance diagnostics in Ads Manager
According to Social Media Examiner, the three diagnostics appear in the default columns view of Ads Manager once an ad has passed the impression threshold. If you don't see them, use the "Columns" dropdown to add quality ranking, engagement rate ranking, and conversion rate ranking. Diagnostics are assessed over the date range you select, so narrowing or widening the window changes the comparison pool.
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 for total value — not raw bid.
According to Segwise, these signals help the system decide "whether your creative deserves more delivery, cheaper impressions, or fewer chances in the auction." The direction 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. For POD sellers running both Meta and Google, understanding how attribution windows interact with delivery costs matters just as much — see our guide to the Google Ads default attribution window for the parallel on that channel.
How to improve your ad relevance diagnostics
There's no button for "raise relevance." You improve the inputs and let the diagnostics follow. According to AdSpyder's 2026 diagnostics guide, the levers map directly to each diagnostic:
- 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. According to Segwise, hides and reports ("people hiding your ad, skipping fast") are what dragged the old score down and still hurt quality ranking today. Avoid clickbait and over-promising.
- Fix the post-click experience. According to AdSpyder, a "simplified landing page, better offer, better tracking + fewer friction steps" directly improve conversion rate ranking even when the ad itself is strong.
- Refresh before fatigue sets in. As frequency rises on a tired creative, engagement falls and relevance slides with it. Watching for ad fatigue keeps your diagnostics from decaying, and steadily working to increase customer engagement is the same work that lifts engagement ranking.
- Audit your tracking setup. Missing or broken conversion events mean Meta has no signal to optimize against, which silently suppresses conversion rate ranking. Confirm your pixel events and Conversions API are firing correctly before reading any diagnostic as an indictment of creative.
The trap: relevance is not profit
Here's what every "relevance score" article leaves out. According to GoDataFeed, "high relevance diagnostics rankings should not be your primary goal" and "high rankings don't guarantee increased results." Relevance lowers your cost per click and cost per result. It says nothing about whether the result made you money.
Say you sell a product with a 50% contribution margin after COGS, shipping, and payment fees. Your break-even ROAS is 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 a strong average ROAS can hide a collapsing marginal ROAS on the last chunk of budget. Those last dollars can lose money while the headline number stays 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. For POD sellers trying to improve overall returns, see our roundup of the best AI tools for improving ROAS on Meta Ads.
According to GoDataFeed, the right way to use diagnostics is as "insight to optimize ads to meet their objectives" — not as a standalone success metric.
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 brings that data into a live data warehouse so the numbers are always current.
On top of that live data sits Victor, an AI employee that reads your Meta ad performance alongside your Shopify margin data and tells you which campaigns are profitable at the margin versus which just look relevant. Victor does not touch your ad account — Meta Ads is a read surface. The writes Victor executes are Shopify-side: repricing worst-margin SKUs, adjusting your free-shipping threshold, creating or updating discount codes, and other moves that directly affect the profit math behind every ad click. Each action shows up as an approval card — old value, new value — and nothing executes until you say yes.
If your relevance diagnostics are green but your margins are murky, that's exactly the gap Victor is built to close. Try PodVector here.
For a broader look at how Victor fits into a POD business, see the PodVector strategy guide or explore the POD seller's guide to generative AI for ecommerce.
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 to your ad-level report.
What is a good ad relevance diagnostic result?
Aim for "above average" on all three. According to Segwise, "above average" means your ad sits in the top 55th percentile or higher against competing ads for the same audience. "Average" is acceptable; "below average — lowest 10%" is the urgent tier. Because the rankings are relative to competitors, "good" shifts with your audience and industry.
How many impressions before I see relevance data?
According to Segwise, your ad must cross 500 impressions before Facebook calculates relevance diagnostics. According to Social Media Examiner, rankings then appear in the default Ads Manager columns view. Below that threshold, don't read too much into an early or missing 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.
Are relevance diagnostics factored into the auction?
The diagnostic rankings themselves are not. According to Meta's announcement, "these ad relevance diagnostics are not factored into an ad's performance in the auction." They are reporting tools that reflect how Meta's auction already scored your ad — the underlying quality signals feed the auction; the diagnostic label is the readout, not the input.
Is relevance the same as ROAS or profit?
No. Relevance lowers your cost per result; it says nothing about margin. As GoDataFeed notes, diagnostics rankings "have no intrinsic value and are not to be used for post-goal optimization." 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.
How do relevance diagnostics relate to ad fatigue?
As frequency builds on a creative, users who have already seen the ad repeatedly are more likely to hide or scroll past it. According to Segwise, negative signals like people "hiding your ad, skipping fast" directly pull quality ranking down. Monitoring frequency alongside your diagnostics is the earliest warning system for fatigue — a drop in engagement ranking before ROAS falls is often the first signal that a creative needs refreshing. For POD sellers managing Klaviyo email alongside Meta ads, a well-timed Klaviyo browse abandonment flow can recover revenue from users who clicked but didn't convert, reducing the pressure to squeeze every last dollar out of a fatiguing ad.