Meta Ad Relevance Diagnostics: what Quality, Engagement, and Conversion Rate Ranking say about your landing page

For years, Meta summed up an ad's perceived quality in a single number, the "relevance score," from 1 to 10. That score has been gone for several years now, replaced by three separate diagnostics that get read independently of one another. Most campaign managers glance at them at best, or only check when cost per result climbs with no obvious explanation. That's a mistake: read together, these three diagnostics say precisely where the problem sits — in the creative, in the targeting, or in the landing page — and stop you from reworking an ad that was already working fine to compensate for a page that isn't converting.

Three diagnostics, three different questions

These metrics live in Ads Manager, at the ad level, under the "Relevance and match" columns. Each one compares that ad's performance against other advertisers targeting the same audience at the same time, and answers a specific question:

• Quality Ranking — the perceived quality of the ad among the people who see it: negative feedback signals like hiding, reports, sensationalized language, or engagement bait.
• Engagement Rate Ranking — expected engagement (likes, comments, shares, saves) compared to competing ads for the same audience: a pure creative-resonance signal.
• Conversion Rate Ranking — expected conversion rate compared to other ads optimizing for the same goal, on the same audience. It's the only one of the three that depends almost entirely on what happens after the click.

A below-average ranking on just one of the three doesn't carry the same weight as a poor score across all of them: reading them together is what actually isolates the real cause, instead of guessing.

Cross-referencing all three to isolate the real cause

Below-average quality and below-average engagement, together, almost always point to a creative problem: the visual or the message isn't landing with the targeted audience, regardless of what page sits behind it. Conversely, average-or-above quality and engagement combined with a below-average conversion ranking isolate a problem that sits after the click — the landing page or the offer itself, since the ad is clearly doing its job of grabbing attention and earning the click.

That last combination is the one that trips up the most advertisers: they see a decent CTR, a creative that "performs" on paper, and wrongly conclude the problem must be targeting or bidding. Conversion rate ranking exists precisely to cut that misdiagnosis short.

The three most common causes behind a poor Conversion Rate Ranking

1. The page's message doesn't deliver on the ad's promise

An ad that leads with a discount, a delivery timeframe, or a specific benefit needs to land on a page that repeats that exact message on the very first screen — not a generic homepage where the visitor has to go hunting for the offer they just saw. Our article on message match between ad and landing page walks through how to check and fix that consistency point by point.

2. The page is slow or awkward on mobile

Meta traffic is almost entirely mobile, and often picked up mid-scroll rather than during active search — tolerance for waiting is even lower there than on a Google search. A page that takes more than two or three seconds to render its main content, or whose elements shift around while loading, loses a chunk of traffic before the message ever gets a chance to land — see our guide on landing page load speed.

3. The offer isn't understandable within a few seconds

A visitor arriving from an Instagram or Facebook feed doesn't carry the active intent of someone typing a query into Google: they need to be reminded, within a few seconds, of the exact reason they clicked, without wading through a wall of text. An unusually high bounce rate on that kind of traffic is often the first visible symptom of this — our article on landing page bounce rate explains how to read it correctly depending on the traffic source.

What the research confirms about ad-to-page congruence and bounces

This link between ad-page match and post-click behavior isn't just practitioner intuition. A study by LaBrecque, Voorhees, Khodakarami and Fombelle (2024), "Native advertising effectiveness: the role of congruence and consumer annoyance on clicks, bounces, and visits" (Journal of the Academy of Marketing Science), shows that congruence between an ad and the destination it leads to directly shapes bounce rate and the number of genuinely useful visits generated, independent of how many clicks the ad pulls in. In other words, a creative that grabs attention without leading to a coherent page produces clicks that are expensive and largely unproductive — exactly the scenario Meta's conversion rate ranking is built to catch.

On the search-engine side, a research team showed that a landing page's bounce rate can be reliably predicted from signals extracted from the page itself — content, structure, how closely the query terms match the visible text — without needing individual behavioral data at all: Dolma, Kalani, Agrawal and Basu (2021), "Improving Bounce Rate Prediction for Rare Queries by Leveraging Landing Page Signals" (Web Conference 2021). That's more or less the same principle behind ad platforms' relevance diagnostics: the page itself carries measurable signals of its own ability to convert, independent of whichever creative led the visitor there.

Fixing a below-average Conversion Rate Ranking

• Build one landing page per ad angle instead of sending every creative to the same generic product page: each tested angle deserves a page that repeats its own message.
• Reproduce the ad's exact offer — same discount, same timeframe, same headline benefit — on the very first screen, without forcing the visitor to scroll to find it.
• Cut friction in the form or checkout flow for cold traffic, which doesn't carry the same intent as a visitor who came back to the site on their own.
• Test load speed under real 4G conditions, not just office Wi-Fi, since that's where most Meta traffic actually happens.
• Let the campaign run a few days after fixing things before drawing conclusions: Meta recalculates these rankings continuously, but needs enough impression volume to stabilize a new score.

A concrete case: a Meta campaign for an e-commerce product

Take a DTC brand running an Instagram ad to its product page, built for instance on a template like single-product e-commerce. If the ad leads with free 48-hour shipping, that detail needs to show up in the page's visible headline, not buried in small print in the terms and conditions. If the conversion rate ranking stays below average despite a creative that engages well, the likeliest cause isn't the targeted audience but the gap between the promise seen in the feed and what the page actually confirms in the first few seconds.

That's also why it pays to start from a page structure built around a single message rather than retrofitting a multi-product page for every new campaign: the 10 LanderKit templates are built as single-offer pages, static and fast — exactly the format a Meta conversion ranking rewards over a generic page shared across ten different creatives.