What is ad relevance?
Ad relevance is how closely your ad matches the intent behind the query, the interests of the audience, and the context it appears in. In Google Ads it is also a literal, named metric: one of the three components of Quality Score, reported per keyword as Below average, Average, or Above average.
That distinction matters, because "ad relevance" means two different things depending on where you are standing:
- In Search, it is a scored diagnostic. Google compares your keyword to your ad text and grades the match. It feeds Quality Score, which feeds Ad Rank, which decides both your position and what you actually pay per click.
- In Display and PMAX, there is no ad relevance column. Nothing grades it for you (why Display has no Quality Score). Relevance is decided by whether your ad reached the right person, in a context that supports the message, on a page worth appearing on at all.
Most guides stop at the first one, because it has a number attached. The second is where the money usually leaks.
How Google scores ad relevance in Search
The Ad relevance column lives in your keyword view. Add it via the Columns menu under Quality Score. Three possible values:
| Status | What it means | What to do |
|---|---|---|
| Above average | Ad text matches the keyword's intent | Nothing |
| Average | Acceptable, not sharp | Worth a pass if the keyword matters |
| Below average | Google sees a mismatch | Fix the ad group, not the ad |
"Below average" almost never means the copy is badly written. It means the ad group is too broad, with one ad trying to serve keywords that want different things. The fix is nearly always to split the ad group so each ad answers one intent, rather than to rewrite the ad in place.
You can check this for any ad group with the free ad relevance checker. Paste the keywords and ad copy and it shows which keywords the ad does not reflect and whether the group should be split.
Does ad relevance affect what you pay?
Yes, in Search. Ad relevance feeds Quality Score, and Quality Score feeds Ad Rank. Two advertisers bidding the same amount do not pay the same amount: the one with better relevance can hold a higher position at a lower cost per click. This is why relevance is a cost lever, not a vanity metric.
In Display and PMAX there is no equivalent score, so nothing tells you when relevance is failing. You find out from your conversion rate, or you do not find out at all.
Why most guides miss the biggest lever
Most guides on ad relevance focus on audience targeting and creative. Both matter. But placement quality is the underrated factor, and it is the one you have the most direct control over.
An ad can match its keyword perfectly and still appear next to scraped content on a site built to farm ad revenue. Nothing in the Quality Score will tell you that happened.
Why Placement Relevance Matters
An ad for a premium product that runs on a clickbait MFA site inherits the environment's credibility (or lack of it). Even if the targeting is perfect, the surrounding content affects how users perceive the brand.
The inverse is also true: a well-placed ad in a relevant, high-quality environment outperforms the same ad on low-quality inventory, often significantly.
The Components of Ad Relevance
Audience relevance. Are you reaching people who are likely to care about your offer? Demographic, behavioral, and contextual signals all contribute.
Creative relevance. Does the ad message speak to something the audience actually cares about? Generic creative underperforms specific creative at every budget level.
Placement relevance. Is the site a place your target audience actually spends meaningful time? Does the surrounding content reinforce or undermine your message?
Practical Steps to Improve Relevance
Review placement reports regularly. Pull your Google Ads placement report monthly. Look at which sites your budget went to. You will find surprises.
Build exclusion lists. Identify site categories that do not fit your brand: politically extreme content, adult material, MFA sites, gambling. Apply exclusions at the account level.
Use inclusion lists for high-value environments. If you know which publications your audience reads, add them to a managed placement list and allocate budget directly.
Measure beyond CTR. Bounce rate, time on site, and conversion rate by placement tell you which sites are actually working. CTR alone is a weak signal, because MFA sites often have high CTR from accidental clicks.
Contextual targeting gets you to the topic, not to the site
Contextual targeting matches ads to content rather than to a person's behaviour, and it comes in two forms. Keyword-based targeting places your ad on pages containing the words you chose. Topic-based targeting places it across pages covering a theme, which reaches further while keeping the subject roughly right.
Both stop at the same boundary. A contextually matched ad on a made-for-advertising site is still a bad placement: the keywords are there, the topic technically aligns, and the environment is still high ad density, thin content and an audience that is not reading.
The division of labour is worth stating plainly, because it is the part most guides skip:
- Contextual targeting gets you to the right topic.
- Placement analysis filters out the bad sites within that topic.
They are complementary, not alternatives. Apply the targeting, then audit what it actually bought.
What changes when the targeting is automated
Hand the decision to automated bidding and the gap moves somewhere harder to see. The platform optimises for what it can measure, not for what matters to your business. A traffic campaign can buy cheap clicks that bring no useful visitors. A lead campaign can report a low cost per acquisition while the leads themselves are weak.
Broad automated targeting can also push budget toward bot traffic, click farms and MFA inventory while the reported metrics improve. Click-through rate goes up, cost per acquisition goes down, and the placements underneath are junk. Nothing in the platform's own reporting tells you which websites the ads ran on or whether anyone real was reading.
What to do depends on the campaign:
- Product sales with clean conversion tracking: let the automation work. The signal is good enough for it to learn from.
- Awareness, traffic and lead generation: add controls. Pull the placement report, audit it for MFA sites and bot-heavy domains, and apply exclusions at account level so one list covers every campaign.
The point is not to fight automated targeting. It is to give it better inputs, and a clean placement list is among the highest-leverage inputs available.
Relevance is a system
Get the targeting right, get the creative right, and get the placement environment right. All three need to be working together, and only the first one has a score attached, which is why the other two get neglected.
If you want to see the placement side of your own account, pull your placement report and run it through the free placement exclusion list generator. It gives you a spend breakdown and a ready-to-paste exclusion list, in your browser, with nothing uploaded.