The Landing Pages Report in Google Ads 2026: Where Paid Traffic Dies After the Click
Account optimisation almost always stops at the click: keywords, bids, ads, audiences. Everything after the click feels like somebody else’s territory — and that is where half the budget quietly goes. It is not actually invisible: the landing pages report shows performance broken out by the URLs your clicks actually land on, including addresses you never specified in a campaign.
Below: where the report lives, how “landing pages” differ from “expanded landing pages”, which columns to read and in what order, eight recurring findings with a concrete action each, and a 30-minute review routine. Every threshold is a reference point — compute your own from your own account.
Where it lives and what it shows
Path in the UI: Campaigns → Insights & reports → Landing pages. The report aggregates performance not by campaign or ad group but by the pages traffic arrived on. It is the only place in the account where data is grouped that way, which is exactly why it surfaces problems ordinary reports hide.
The classic example: a campaign looks fine, CPA within target. Inside it, four ad groups point at four pages, one of which converts at zero while absorbing a third of the traffic. Averaged into a campaign row it is invisible. In the landing pages report it is the first thing you see.
Two modes: standard and expanded
The report has a “View expanded landing pages” toggle, and the difference matters:
- Landing pages — aggregated data for the URL set in the ad or ad group.
- Expanded landing pages — the specific final URLs users actually reached after the click.
The second mode is the main reason to open this report in 2026. Wherever the system picks a destination for you — final URL expansion in Performance Max and AI Max, dynamic targets, sitelinks — the real address can diverge sharply from the one you set. How to constrain that picking is covered in final URL expansion in Performance Max and AI Max; this report is where you first see what it actually chose.
Columns: what to read and in what order
Some columns are specific to this report; the rest come from the standard metric set through column customisation.
| Column | What it actually means |
|---|---|
| Mobile-friendly click rate | The share of clicks that landed on pages passing Google’s mobile-friendly test. Not “does it load fast” but “is it usable on a phone at all” |
| Valid AMP click rate | The share of clicks that loaded as valid AMP pages. A dash means there is no AMP markup — not an error |
| Clicks / Impressions / CTR | Raw volume. For Search, when several sitelinks show in one impression, only the unique ad impression is counted |
| Cost | What this page actually cost you. The column to sort by when you start |
| Conversions and conversion rate | The only metric the exercise exists for. Never read without cost beside it |
| Conversion value | With values configured, pages rank by money rather than by count |
A reading order that saves time: sort by cost descending, then look at conversions and conversion rate, and only then at the technical columns. Technical metrics explain why; money is what finds the problem.
The report segments by time, device, conversion type and network. Device is the mandatory first segment: the most common finding here is a page that converts acceptably on desktop, not at all on mobile, and looks merely “okay” on average.
The landing pages report: eight recurring findings
1. High spend, zero conversions
The most expensive finding. Before blaming the page, check whether it fires a conversion at all: a zero often means a missing tag rather than a bad page. How to tell those apart is covered in why Google Ads and GA4 conversions never match. If tracking is healthy, split the diagnosis: either the traffic is wrong for the page, or the page does not deliver what the ad promised.
2. An expanded URL you never set
The system picked a destination itself. Sometimes well, sometimes onto an About page, a blog post or a product that is out of stock. The fix depends on the campaign: exclude unwanted sections, turn expansion off, or add the missing landing page to your structure properly.
3. Low mobile-friendly click rate
A value well below 100% with live mobile traffic is technical debt that is costing money right now. It is a separate problem from speed — a page can load fast and still be unusable on a phone. Speed has its own treatment in landing page speed and Core Web Vitals for paid traffic.
4. Same offer, two pages, very different CPA
The most useful finding for scaling: you already have an internal benchmark. Work out what the better page does differently — form length, number of fields, price visible, trust block — and port it across. Verify the port by measurement rather than by feel; the correct setup is in experiments in Experiment Center.
5. Traffic landing through a redirect
Expanded mode shows the final address differing from the one you set through a redirect chain. Every hop is lost fractions of a second and a risk of losing tracking parameters. If the click identifier is dropped, the conversion never reaches the account at all; the mechanics are in GCLID, GBRAID and WBRAID.
6. Rows multiplying because of URL parameters
One product split across a dozen rows by UTM tails, session IDs and sort orders. That is not only unreadable — it can mean traffic is landing on URLs you do not control. Normalise your tagging and check whether the site itself is generating stray parameters.
7. A page collecting clicks that is not in your campaign structure
Typical for campaigns with automatic destination selection and for dynamic targets. Either build a proper ad group around it with relevant ads, or exclude it. Leaving it is the worst option: the ad promises one thing and the page shows another.
8. Good conversion rate, barely any traffic
The inverse problem: not a fix, a scale-up. Check which ad groups feed it and whether they are budget or rank limited — impression share and auction insights tells you which ceiling you are hitting.
What the report does not show
- No on-page behaviour. Scrolls, element clicks, abandoned forms — that is analytics, not the ad account.
- No causes. The report says “this does not convert”, never “because the form has nine fields”.
- Nothing downstream. Lead quality and payment live in the CRM.
- A dash is not always a problem. No mobile clicks, no mobile-friendly value; no AMP markup, a dash in the AMP column.
- It does not replace looking. A page with bad numbers needs to be opened on a phone and walked through to conversion by hand.
The landing pages report answers “where is the money going”, and almost never “why”. The second half of the answer always lives outside the ad account.
A 30-minute review
- Pick 28–90 days. Shorter and individual URLs have no statistics; longer and you blend old versions of pages.
- Sort by cost. The top ten rows usually carry 70–80% of spend — a reference worth checking against your own account.
- Add columns: cost, conversions, conversion rate, conversion value, mobile-friendly click rate.
- Segment by device. Look for rows where mobile conversion is a multiple worse than desktop.
- Switch to expanded mode. Check whether any actual destination was never in your plan.
- Tag three categories: “expensive and dead”, “unexpected URL”, “converts well, starved of traffic”.
- One action per row, not a list of ideas: exclude, redirect, fix, scale.
- Date the review and come back to the same rows in a month.
A worked example
Services account, 60 days, $9,400 spend, 212 conversions, average CPA $44. Campaign view looks even. Open the report, sort by cost.
| Page | Spend | Conversions | CPA | Conv. rate |
|---|---|---|---|---|
| /services/audit/ | $3,100 | 96 | $32 | 5.8% |
| /services/retainer/ | $2,600 | 71 | $37 | 4.9% |
| /services/consultation/ | $2,050 | 12 | $171 | 0.9% |
| /blog/how-to-choose-an-agency/ | $980 | 3 | $327 | 0.4% |
| /contact/ | $670 | 30 | $22 | 7.1% |
What a minute with this table shows, and the campaign report does not:
- Row three eats 22% of budget at five times target CPA. On inspection, its form posts to a separate thank-you URL that was never tagged — part of the result simply never arrives. That is a measurement job, not a page job.
- Row four is a blog post that collected nearly a thousand dollars. It was never set as a landing page; automatic selection chose it. Fix: exclude the blog section from selection and route that demand to a service page.
- Row five has the best conversion rate in the account and almost no volume. Its feeding ad group is budget limited. Not a problem — a growth point.
- Rows one and two are your internal benchmark. A gap between 5.8% and 4.9% on the same offer is a reason to compare the form and the pricing block, not a reason to panic.
Outcome: one measurement bug, one settings exclusion, one scale-up and one test hypothesis. Four concrete actions out of one table in half an hour — none of them visible in the campaign report.
How it looks in three kinds of account
| Account type | What to look for first |
|---|---|
| E-commerce | Out-of-stock product pages; categories instead of product pages; sort-parameter URLs smearing the data |
| Lead generation | Conversion spread between similar service pages; zeros where the form posts to a separate thank-you page |
| Local business | Location and contact pages — frequently the best conversion rate and the least traffic |
Connecting it to the rest of your optimisation
| Finding | Where it goes next |
|---|---|
| Irrelevant traffic to a decent page | Search terms report and negative keywords |
| Relevant traffic to a weak page | Page work: form, offer, speed |
| Unexpected expanded URLs | URL expansion settings and campaign structure |
| Different conversion on the same offer | An experiment, then porting the change |
| Zero conversions with live traffic | Check tracking, do not touch the page yet |
| Good page, small volume | Budget, bids, impression share |
Worth saying out loud: the actions in the right column sit in different teams and cost wildly different amounts. Excluding a section from automatic selection is five minutes in settings. Fixing a conversion handoff is a day with a developer. Rewriting a page is weeks. So prioritise findings by the ratio of likely saving to cost of fixing, not by raw spend: two cheap fixes shipped this week usually beat one expensive fix deferred to next quarter.
One note on automated bidding: Smart Bidding optimises toward the conversions it sees and will gradually starve pages that do not convert. It does this slowly, and only if conversions are reported correctly. The report matters because a human spots these things faster than the algorithm and can fix the cause rather than the symptom.
Common misconceptions
“Quality Score is fine, so the page is fine.” Landing page experience measures relevance and expectation matching, not the ability to sell. A page can be perfectly relevant and convert at zero because of its form or its price.
“This is an e-commerce report.” The opposite: in lead generation, where there are few pages, every row carries more weight, and the conversion spread between two superficially similar pages is usually wider.
“It does not work for Performance Max.” It does, and expanded mode is most valuable precisely there — it is the fastest way to see which addresses the system chose on its own.
“Zero conversions means a bad page.” Half the time it means a missing or broken tag. Measurement first, conclusions second.
When the report is useless
- All traffic goes to one page. The report collapses to one row and tells you nothing until you have several destinations.
- Conversions are not configured. You get clicks and cost, which show spend but support no decision.
- Less than a month of data. Over two weeks the variance per URL is large enough that any conclusion is noise.
In all three cases the order is reversed: measurement and structure first, page analysis after.
Making it a routine
- Monthly — a full review on the schedule above, with actions logged wherever you log other account changes.
- After every site release — check expanded URLs: redirects and URL structure changes surface here first, usually within days.
- Before a seasonal peak — a dedicated pass, because the cost of a broken page rises with bids during a peak.
- Split the roles: the ads side owns traffic relevance and selection settings, the site team owns the page. The classic failure is a finding stranded between the two for months.
A practical trick: bring the site team a row with spend and CPA next to a benchmark row from the same account, rather than “the page converts badly”. The conversation about priority takes a minute instead of a week.
If the review reveals something broader than one page — values not being passed, or conversions going missing — work through the Google Ads account audit checklist. And before drawing conclusions about a page from account numbers, make sure Google can read it at all: a landing page blocked from the crawler produces strange metrics unrelated to its quality — see AdsBot and landing page crawlability. If competitors in your niche have just changed their offers, that shows up in your conversion rates too; what they are running is visible through the Google Ads Transparency Center. For a guided pass over your own account, PPC Rebels runs Google Ads training, and the sequence above is kept in the account playbook.
FAQ
Where is the landing pages report?
Campaigns → Insights & reports → Landing pages. The toggle for expanded landing pages sits in the same view.
How do expanded landing pages differ from standard ones?
Standard shows the URL you set; expanded shows the address the user actually reached. The difference comes from automatic destination selection, dynamic targets, sitelinks and redirects.
Why is the mobile-friendly column showing a dash?
The page received no mobile clicks, so there is nothing to compute. It is not a warning sign.
What does a dash in the AMP column mean?
That the page has no AMP markup. For most sites that is the normal state, not a fault.
Can I add cost and conversions to this report?
Yes, standard performance metrics are added through column customisation. Sorting by cost is the easiest way to start.
What date range should I use?
28 to 90 days as a reference. Shorter and individual URLs lack statistics; much longer and different versions of a page get blended.
How many conversions do I need before comparing two pages?
As a reference, at least a few dozen each. Below that, the difference between pages usually sits inside random variation.
High spend and zero conversions — what first?
Check whether the page fires a conversion at all. Only once measurement is confirmed healthy should you look at traffic relevance and the page itself.
Does the report work for Performance Max?
Yes, and expanded mode is especially useful there: it shows which destinations the system picked by itself.
Why has one page split into many rows?
URL parameters: tags, session IDs, sort orders. Normalise the tagging or the data stays smeared.
Does this replace web analytics?
No. It shows which pages cost money and which convert, but not what happens on the page or why.
How often should I open it?
Monthly for a stable account, plus after every site change, after launching campaigns with automatic destination selection, and before a seasonal peak.
