What "relevance score" means now
Meta retired the old 1-to-10 relevance score and replaced it with three ad relevance diagnostics: quality ranking, engagement rate ranking, and conversion rate ranking. Each is rated Above Average, Average, or Below Average.
The key word is ranking. Per Meta's ad relevance diagnostics documentation, each metric compares your ad against other ads competing for the same audience with the same optimization goal. So "high" always means high versus your competition, not high on an absolute scale.
That framing matters for diagnosis. If your score is high, you are winning a local contest — you are simply a better answer than the other advertisers bidding for the same eyeballs right now.
Why is my relevance score high? The three real reasons
When people ask "why is my relevance score high," they are usually seeing green diagnostics and want to know what earned them. Three things drive it.
1. Your creative earns positive feedback, not negative
Quality ranking leans heavily on feedback signals. Likes, shares, saves, and watch time push it up; hides, "I don't want to see this," and spam reports drag it down.
A high quality ranking means people who see your ad react well and rarely try to make it go away. Your hook, format, and offer are landing with the audience Meta chose.
2. Your estimated action rate beats the field
Meta does not simply award impressions to the highest bidder. It runs an auction on Total Value, described by practitioners as roughly Bid × Estimated Action Rate + Ad Quality.
Estimated Action Rate (EAR) is Meta's prediction that this user takes your optimized action — roughly estimated click-through rate times estimated conversion rate. A high engagement or conversion ranking means your EAR is strong, so you can win the impression against someone bidding more and still pay a lower price for it.
3. You match the audience Meta actually found
Conversion rate ranking asks whether people who click go on to do what you optimized for. When it reads above average, your ad, landing page, and offer form one coherent path.
Since Meta now treats creative as the primary targeting signal, a strong creative effectively is good targeting. The system uses your hook to find the people most likely to convert, which reinforces the score. Our guide to why a relevance score comes back low walks the mirror image of this.
What a high score actually buys you
A high relevance score generally means a lower cost per result, because a high EAR and positive quality signals push your CPM down in the auction.
That is real and worth having. Cheaper impressions mean more reach per dollar and more room before rising costs squeeze you. If you want to push those diagnostics higher on purpose, our playbook on how to improve your relevance score is the practical next step.
But notice what the score does not measure: your margin. Relevance is about delivery efficiency inside Meta's auction. It says nothing about your product cost, shipping, or fees — the things that decide whether a cheap click becomes a profitable order.
The trap: high relevance score, thin profit
Here is where advertisers get burned. A high relevance score and a healthy-looking Return On Ad Spend (ROAS) can both be true while your last dollars of spend lose money.
Two numbers explain it: break-even ROAS and marginal ROAS.
Break-even ROAS is set by your margin, not your ad
Break-even ROAS is pure arithmetic: 1 ÷ contribution margin, where contribution margin is the share of revenue left after variable costs (product, shipping, transaction fees, pick-and-pack) but before ad spend.
Say you sell a $50 product with a 50% contribution margin. That leaves $25 per order to cover ads and profit, so your break-even ROAS is 50 ÷ 25 = 2.0x. Below 2.0x you lose money no matter how green your relevance diagnostics look. If your margin were thinner — say 30% — break-even climbs to 1 ÷ 0.30 = 3.33x, and paid acquisition gets hard fast.
A high relevance score can make ROAS look fine while sitting under your break-even line. The auction rewarded your ad; your P&L did not.
Marginal ROAS is where scaling breaks
Now scale that winner. The auction serves your cheapest, most-responsive audience first, so every extra dollar reaches a slightly worse slice. Your average ROAS can stay green while the marginal return on new spend quietly falls.
Say last week you spent $3,000 and made $12,000. This week you pushed to $5,000 and made $13,200. Average ROAS still looks strong at $13,200 ÷ $5,000 = 2.64x. But the marginal ROAS on the new spend is ($13,200 − $12,000) ÷ ($5,000 − $3,000) = $1,200 ÷ $2,000 = 0.6x. Your last $2,000 lost money, even though relevance stayed high the whole time.
That is the core lesson: relevance tells you the auction likes your ad; only per-order profit and marginal ROAS tell you whether to spend more. Scaling decisions live on the marginal number, a point we unpack in profit-driven marketing and in the broader profitable ad scaling guide.
How to keep a high relevance score
A high score is not permanent. It decays as an audience sees the same creative too many times — frequency climbs, feedback sours, and quality ranking slips.
Protect it with a few habits. Refresh creative before it fatigues rather than after; watch click-through rate and frequency together, since a falling CTR against rising frequency is the early fatigue signal. Keep your landing page fast and consistent with the ad, because conversion rate ranking punishes a mismatch.
And keep the audience wide enough to feed the ad. A narrow audience saturates quickly, and a saturated audience is where a once-high score goes to die.
Where per-order profit fits in
The blind spot in every relevance metric is money. Meta can tell you your ad is well-received; it cannot tell you a $50 order netted $6 after product, shipping, and fees.
That gap is what PodVector exists to close. It connects Shopify, Meta Ads, Google Ads, Printify, and Printful, then computes your true per-order profit — so a green relevance score gets checked against what each order actually keeps.
Victor, our AI operator, reads that live data and flags where a high-relevance, high-ROAS campaign is thin on real margin. He proposes moves and, with your approval, executes the Shopify-side ones — think free-shipping thresholds or bundles that lift average order value. Victor does not touch your ad account, and PodVector is not a dashboard; it is the profit layer that tells you whether your cheap, relevant clicks are worth scaling.
One high-leverage lever is average order value, because a one-click post-purchase upsell on Shopify adds margin at zero extra acquisition cost — which lowers the break-even ROAS your relevant ads have to clear.
FAQs
Is a high relevance score always good?
Mostly yes, but with a caveat. High diagnostics usually mean a lower cost per result, which is genuinely useful. The caveat is that relevance measures auction performance, not profit — a high score can sit above a campaign that loses money once product cost, shipping, and fees are counted.
Does a high relevance score guarantee a lower cost per result?
It strongly tends to. Because Meta's auction ranks on Total Value — bid times estimated action rate plus ad quality — a high EAR and clean quality signals let you win impressions at a lower CPM than higher bidders. It is a tendency baked into the auction, not a promise, and market factors like seasonal auction density can raise everyone's CPM regardless.
Why is my relevance score high but sales are flat?
Because relevance and conversion measure different things. A strong quality and engagement ranking means people react well to the ad, but if conversion rate ranking lags, the clicks are not turning into orders — often a landing page, price, or offer mismatch. Check whether your conversion ranking is the weak link and whether tracking is reporting sales accurately.
Can my relevance score be high while I lose money?
Yes, and it is common. Relevance ignores your contribution margin entirely. If your break-even ROAS is 2.5x and your campaign runs at 2.0x, you lose money on those orders even with all three diagnostics above average.
How is relevance score different from ROAS?
Relevance score (now the three diagnostics) rates how well your ad performs inside Meta's auction versus competitors. ROAS is revenue divided by ad spend. Neither one accounts for cost of goods, shipping, or fees, which is why per-order profit is the number that actually decides whether to scale.
Will my high relevance score last?
Not on its own. Scores erode as creative fatigues and frequency climbs on the same audience. The fix is a steady cadence of fresh creative and an audience large enough that it does not saturate before the ad does.