Landing Page Experience: The Changes That Actually Move It
A practitioner's guide to raising the Below Average landing page rating, what matters, and what is a waste of an afternoon.
The short version
- Fix the intent gap between query and page first; it moves the rating more than speed or design.
- The per-keyword label reflects intent mismatch for that query, but the underlying page quality is assessed at a page or account level, so treat 'same URL, different rating' as a signal, not proof.
- Use PageSpeed Insights field data and Google's Core Web Vitals bands; fix red and amber metrics, leave green ones alone.
- Kill entry interstitials and pop-ups on paid traffic; make the offer and contact info reachable without dismissing anything.
- Change one thing at a time and give the page real traffic, because the rating refreshes slowly and you cannot isolate causes otherwise.
If your landing page experience is rated Below Average, the three changes that move it are: match the page content to the query intent the keyword is buying, make the page usable and fast on mobile, and remove friction and interstitials that block the content a user came for. Cosmetic tweaks, swapping a hero image, or adding trust badges rarely move the rating. Fix the intent gap first, because that is the single lever that correlates most with the label.
Find which pages are actually rated, then read the label correctly
Start in the UI. Add the Landing page experience column to your keyword view (Columns > Modify columns > Quality Score > Landing page exp.). You only see this for keywords that have accumulated enough impressions, and the value shows as Above average, Average, or Below average. Sort by Below average and look at which URLs those keywords point to. That is your work queue.
Here is the important nuance, and it is where a lot of advice overreaches. Google displays this rating alongside the keyword, which makes it tempting to treat it as a pure per-keyword score. It is not cleanly that. Google has stated landing page quality is often assessed at a page or account level, not purely per keyword. What the per-keyword label reflects is the gap between that query's intent and what the destination page delivers, layered on top of the page's underlying quality. So when the same URL shows Below average for one keyword and Average for another, the most likely explanation is an intent mismatch for the first query, not that Google literally re-rated the page per keyword. Treat the 'same URL, different rating' pattern as a diagnostic signal of intent mismatch, not as documented proof of keyword-level page scoring. It is inference, and useful inference, but not fact.
That distinction changes what you do. If one page is Below average across many keywords, fix the page. If one page is Below average for a handful of specific queries but fine for others, the problem is almost certainly that those queries are landing on the wrong page or expect content that is not there.
The intent gap is the biggest lever
The rating is Google's read on whether a person who clicked your ad for a given query finds what they expected, quickly, and can act on it. The most common failure is not a slow page. It is a page that answers a different question than the query asked.
Concrete example: a keyword like 'commercial dishwasher repair near me' pointing at a generic 'Our Services' page will underperform a page that says commercial dishwasher repair in the headline, lists the service area, and has a phone number and a request-service form above the fold. The query is transactional and local; the page needs to confirm both facts within the first screen. If your ad promises 'same-day repair' and the landing page never mentions same-day, that is a relevance gap the system can detect through behavior and content.
Decision rule: for every Below average keyword, open the destination URL and read it as if you just searched that exact phrase. If the main keyword's intent is not confirmed in the headline and the primary call to action within the first screenful, that is your fix. Build or route to a page that matches. This is usually a bigger win than any speed optimization.
A related trap: routing all ad traffic to the homepage. Homepages are built for navigation, not for a specific query's intent, so they reliably rate worse than a dedicated page. If you are sending multiple ad groups to one broad page, split the destinations.
Mobile usability and speed, with honest thresholds
Most paid clicks are mobile, and the rating weights the mobile experience heavily. Run your top Below average URLs through PageSpeed Insights and look at the field data (real user metrics) rather than only the lab score. Field data reflects what actual visitors experienced, which is closer to what Google uses.
The metrics that matter are Core Web Vitals: Largest Contentful Paint (how fast the main content appears), Interaction to Next Paint (how fast the page responds to a tap), and Cumulative Layout Shift (whether the page jumps around as it loads). Rather than chase a specific second-count I cannot source for ad ratings, use Google's own published Core Web Vitals thresholds in PageSpeed Insights: if a metric is in the red or amber band, it is worth fixing; if it is green, your time is better spent elsewhere. The biggest real-world speed wins are compressing hero images, deferring third-party scripts (chat widgets, tag managers loading too much), and removing render-blocking resources.
Usability failures that drag the rating: tap targets too small or too close together, forms that are painful on a phone, and content hidden behind horizontal scroll. Fix these before you obsess over shaving milliseconds off an already-green page.
Friction, interstitials, and transparency
Google explicitly penalizes experiences that make the user's goal harder to reach. The usual suspects: full-screen interstitials or pop-ups that cover the content on arrival, aggressive cookie or subscription modals, and pages that redirect or auto-download. If a user has to dismiss something before they can read the page, assume it hurts.
Transparency matters too. Pages that clearly state what the business does, make contact information easy to find, and do not bury the actual offer tend to rate better. This is less about trust badges and more about not looking deceptive: if the ad sells a product and the page is a thin affiliate redirect, that is the kind of thing the system is built to catch.
Decision rule: delay any pop-up until the user has engaged (scroll or a time delay), or move the offer inline. Do not show an entry interstitial on paid traffic at all if you can avoid it.
What does not move it, and how long to wait
Things that feel productive but rarely change the rating: swapping stock photos, changing button colors, adding more testimonials, or rewriting copy that already matches intent. These can lift conversion rate, which is worth doing, but conversion rate is not the landing page experience input. Do not confuse the two.
The hardest part is feedback latency. The rating does not update the moment you publish a change. It recalculates as the page accumulates new impressions and visits, so you may wait days to weeks to see Average or Above average appear. Resist the urge to make five changes at once and then not know which one worked. Change the intent match first, ship it, and give it real traffic before touching anything else.
Honest limit: Google does not expose the exact inputs or weights behind the rating, and it does not tell you which specific factor dragged a given page down. Everything here is informed inference from how the system behaves and what Google documents. You are debugging a black box, so change one variable at a time and read the label as it refreshes.
A note on wasted spend while you fix pages
Landing page fixes take time to propagate, but the traffic hitting those pages costs money today. While you rebuild intent-matched pages, tighten the queries reaching them. A query that never matches your page intent is both a wasted click and a drag on relevance.
Pull Campaigns > Insights and reports > Search terms, and cut the terms that bring intent your page will never satisfy. Negative keywords stop the bleeding immediately, which buys you time to do the page work properly. QueryCut automates surfacing those negatives from your search terms if you do not want to grind through the report by hand, but the manual path works fine for a small account.
The point is sequencing: stop the irrelevant traffic now, fix the intent match on the page next, and let the rating catch up. Do not wait for the rating to improve before you cut the obviously bad queries.
Questions people also ask
How long does it take for landing page experience to update after I fix the page?
There is no fixed timer. The rating recalculates as the page gathers new impressions and visits, so a low-traffic keyword can take weeks while a high-traffic one updates faster. Ship one change, let it accumulate real traffic, then read the label. Making several changes at once leaves you unable to tell which one worked.
Does landing page experience affect my ad rank and cost directly?
Yes. It is one of the three Quality Score components, and Quality Score feeds into Ad Rank, which affects both your position and what you pay per click. A Below average rating effectively raises your cost to compete. That is why fixing it is worth real engineering time, not just better conversions.
Should I build a separate landing page for every keyword?
No, that is overkill and unmaintainable. Build pages at the level of intent, not the level of individual keywords. Group keywords that share the same intent onto one tightly matched page, and split only when the intents genuinely differ, for example separating a 'repair' intent from a 'buy new' intent. Sending everything to the homepage is the mistake to avoid.
My page is fast and green on Core Web Vitals but still rated Below average. What now?
Then speed is not your problem; intent match almost certainly is. Read the page as if you had just searched the keyword and check whether the headline and primary call to action confirm that exact intent within the first screen. Also check for interstitials, misleading ad-to-page promises, or traffic landing on a page that answers a different question.
Is the landing page rating really per keyword or per page?
The UI shows it next to the keyword, but Google has indicated page quality is often assessed at a page or account level rather than purely per keyword. The per-keyword label is best read as the intent fit between that query and the page, sitting on top of the page's underlying quality. Use it to spot mismatches, but do not assume Google literally re-scores the same URL independently for every keyword.