If you've just opened your Google Ads account and seen "Above average" sitting in the ad relevance column, congratulations - sort of. A high ad relevance score is a good signal, but it is one of the most misunderstood metrics in the platform. Ben sees this every week: advertisers proudly point to a column full of "Above average" labels, then ask why their CPC is still climbing and their conversions are flat.
Here is the honest answer. A high ad relevance score means your ad copy closely matches the keywords in the ad group. That is it. It does not mean your ads are profitable, your landing page converts, your bidding is correct, or your account is healthy. In some cases, an "Above average" relevance score is hiding a deeper structural problem - usually keywords that are too narrow, low search volume, or buyer intent that does not match the offer.
This guide unpacks exactly what ad relevance is, how Google calculates it, why yours is high, what it means for your CPC and Quality Score, and the cases where a high score is masking a weak account. Everything below is based on UK data and Ben's experience auditing 250+ Google Ads accounts.
The short answer
Your ad relevance is "Above average" because the words in your ad copy and headlines closely mirror the keywords you are bidding on, and your ad group is tightly themed. Google's algorithm has decided your ad answers the search query better than most advertisers in your auction. That is a positive signal - it is one of three components of Quality Score - but it is not the same as performance.
Key distinction: Ad relevance measures the match between your ad and your keyword. It does not measure the match between your ad and the user's actual intent, and it does not measure whether the click made you any money.
What ad relevance actually measures
Ad relevance is one of three diagnostic columns Google exposes inside the Keywords tab. The other two are expected click-through rate and landing page experience. Together, these three feed into the Quality Score (1-10) that Google calculates for every keyword in your account.
Specifically, ad relevance answers a single question for Google: How closely does this ad match the meaning of the keyword that triggered it? It is not about persuasion, design, or offer strength. It is purely a semantic match between the ad text and the keyword theme.
If you are bidding on the keyword "emergency plumber London" and your ad headline reads "Emergency Plumber in London - 24/7 Call-Out", Google will score that as highly relevant. If your headline instead reads "Trusted Home Services Since 1998", the ad is still legitimate, but Google considers it less relevant to that specific keyword.
How Google calculates ad relevance
Google has never published the exact formula, but based on Google's own documentation and a decade of agency-side testing, the inputs are well understood:
- Keyword presence in headlines and descriptions - whether the keyword (or close semantic variants) appears in the ad copy.
- Ad group theme tightness - whether all keywords in the ad group share a common theme that the ad addresses.
- Semantic match - Google's natural-language models check meaning, not just exact strings. "Plumber" and "plumbing services" are treated as related.
- Ad strength signals - for Responsive Search Ads, the variety and quality of headline/description assets influence the relevance signal.
- Search query alignment - although ad relevance is technically per-keyword, real-world performance against actual search queries also feeds into the rolling calculation.
The diagnostic is updated on a rolling basis. A new ad will start at "Average" until Google has enough impression data to make a confident judgement - usually a few hundred impressions per keyword.
Below, Average, Above average - what each means
Google reports ad relevance in three buckets only. There is no numeric score exposed for this metric.
| Status | What it signals | Action |
|---|---|---|
| Below average | Ad copy does not closely match the keyword theme. Often caused by overly broad ad groups. | Split the ad group, rewrite ad copy to include the keyword theme. |
| Average | Acceptable match. Most accounts sit here. Not a problem in itself. | Leave alone unless other Quality Score components are weak. |
| Above average | Strong semantic match between ad and keyword. | Maintain. Do not assume this means the account is performing. |
Why your ad relevance is high
If your ad relevance is sitting at "Above average" across most keywords, one or more of the following is true:
1. Your ad groups are tightly themed
This is the most common - and healthiest - reason. You have ad groups built around a single concept (e.g. "boiler installation London") rather than dumping 80 unrelated keywords into one ad group. Tightly themed ad groups make it trivial for ad copy to mirror the keyword.
2. You are using single-keyword ad groups (SKAGs)
SKAGs are out of fashion since RSAs took over, but many accounts still use them - and they almost always score "Above average" on ad relevance because the ad copy literally contains the one keyword in the group.
Watch out: SKAGs frequently produce "Above average" relevance and "Below average" expected CTR at the same time, because the keyword has so little volume Google cannot judge CTR confidently. High relevance, low impressions, no learnings.
3. Your RSA assets are well-written
Responsive Search Ads pin the keyword into headlines and descriptions naturally. If you have written 8-15 distinct headlines that all reflect the ad group theme, Google's RSA assembler will combine them into highly relevant ads.
4. You are bidding on branded or low-competition keywords
Brand terms (your own company name) almost always score "Above average" on relevance because no other advertiser is more relevant for your brand than you are. Same for niche, long-tail keywords with one obvious answer.
5. You are using exact match or phrase match
Tighter match types restrict the queries that trigger your ads, which keeps the keyword-to-ad alignment cleaner. Broad match - especially in 2026's "smart broad" era - tends to drag relevance down because ads are shown against tangentially related queries.
High relevance but no conversions: what's wrong
This is the conversation Ben has on most audit calls. The advertiser opens the platform, points at a column of "Above average" labels, and asks why the account is losing money. The honest answer is that ad relevance does not measure any of the things that determine profitability:
- It does not measure intent. "Cheap plumber London" and "best plumber London" can both be highly relevant to a plumbing ad, but only one of them tends to convert.
- It does not measure landing page quality. A perfectly relevant ad pointing at a slow, generic homepage will still waste budget.
- It does not measure offer strength. If your competitor is offering free quotes within 30 minutes and you're offering "contact us", relevance is irrelevant.
- It does not measure search volume. A SKAG with 12 impressions a month can be highly relevant and contribute zero revenue.
- It does not measure conversion tracking accuracy. Many "non-converting" accounts are actually converting - the tracking is just broken.
If your relevance is high but conversions are low, the diagnosis is almost always one of: weak landing page, mismatched intent in the keyword list, broken or under-counted conversion tracking, or an offer/price that simply isn't competitive.
Ad relevance vs Quality Score - the difference
People conflate these constantly. They are not the same.
Quality Score (1-10) is a numeric score per keyword. It is calculated from three diagnostic components:
- Expected click-through rate
- Ad relevance
- Landing page experience
Ad relevance is just one of those three components, and arguably the easiest to influence. You can have "Above average" ad relevance and still hold a Quality Score of 4 if your expected CTR or landing page experience is "Below average". This is a very common pattern in accounts that have invested heavily in ad copy but ignored landing pages.
Rule of thumb: Aim for "Above average" or "Average" on all three components. A single "Below average" component is enough to drag Quality Score below 6 and inflate CPC by 25-40% in competitive UK auctions.
Does high ad relevance lower your CPC?
Indirectly - yes. Directly - not as much as people think.
Quality Score is the lever that Google uses to discount or inflate your bids. Ad relevance is one input into Quality Score. According to Google Ads Help, moving from a Quality Score of 5 to a Quality Score of 10 can cut your effective CPC by around 50%. But that swing requires all three components to be strong, not just ad relevance.
In practical UK terms, on the WordStream 2024 average Search CPC of £1.05, going from Quality Score 5 to Quality Score 10 might take a click from £1.05 down to £0.50-£0.60. Ad relevance alone would only contribute roughly a third of that improvement.
Where ad relevance has the biggest CPC impact is in Ad Rank. Higher relevance contributes to a higher Ad Rank, which lets you win the auction at a lower bid. But again, this only matters if expected CTR and landing page experience are also strong.
Ad relevance in RSAs and Performance Max
The way ad relevance is reported has changed since the move away from Expanded Text Ads.
Responsive Search Ads: Google still reports the keyword-level ad relevance diagnostic, but it now reflects the average performance across all combinations of headlines and descriptions Google assembled. To keep relevance high in RSAs:
- Write 12-15 distinct headlines per RSA, each reflecting the ad group theme.
- Pin one or two headlines that contain the core keyword in position 1.
- Avoid generic headlines like "Trusted Brand" - they dilute relevance.
- Aim for an ad strength of "Good" or "Excellent" - this correlates with higher relevance.
Performance Max: Ad relevance as a diagnostic does not exist in PMax. Google reports asset performance ratings instead (Low, Good, Best). The closest equivalent of ad relevance in PMax is matching your asset group to a tightly themed audience signal and a clear set of products or services. Mixing unrelated services into one asset group will hurt performance the same way a sloppy ad group hurts relevance in Search.
How to keep ad relevance high without losing intent
The risk of optimising for ad relevance alone is that you end up writing ads that match keywords but don't sell. Here is how Ben balances it in client accounts:
- Start with intent buckets, not keyword lists. Group keywords by what the searcher actually wants - "research", "comparison", "ready to buy" - then build ad groups inside each bucket.
- Theme ad groups around verbs and outcomes, not just product names. "Get a quote", "book today", "compare prices" all signal intent and naturally include keyword-matching language.
- Use ad customisers to dynamically insert location, price, or stock count. This raises relevance and addresses real intent at the same time.
- Pin only the headlines that need to be in position 1 - usually the keyword headline and a primary CTA. Over-pinning kills RSA performance.
- Refresh ads quarterly. Ad relevance can decay as Google's models update. A quarterly review keeps assets fresh and relevance high.
- Match landing page H1 to ad headline. This boosts landing page experience too - a double Quality Score win.
Ben's take from 250+ accounts
In Ben's experience auditing UK accounts across legal, ecommerce, home services, B2B SaaS and finance, ad relevance is one of the easiest metrics to get to "Above average" and one of the most over-celebrated. The accounts that genuinely scale are not the ones with the highest relevance scores - they are the ones with strong intent matching, fast and clear landing pages, accurate conversion tracking, and disciplined negative keyword work.
When Ben starts a 90-Day Revenue Takeover, ad relevance is rarely the first lever pulled. The order is usually: fix tracking, audit search terms, restructure ad groups around intent, rebuild landing pages, then refine ad copy. Relevance tends to climb naturally once the structure is right.
The honest summary: a high ad relevance score is a positive signal, not a victory. Treat it as a hygiene check, not a north star. The accounts that win are the ones treating relevance as the floor, not the ceiling.
