Zombie pages: the organic pages that get visits and never convert
Some pages rank, collect clicks every day and never produce a buyer. How to find them in Search Console and GA4, decide what each one needs, and report SEO by buyers instead of visitors.
- A zombie page earns organic visits and produces nothing. It shows no conversions and no assisted conversions, so its traffic inflates the SEO report without adding a buyer.
- Finding them takes three reports joined on the page URL. Search Console for clicks and queries, the GA4 landing page view for same-session key events, and a GA4 exploration for later-visit conversions.
- The queries decide the fix. Wrong intent cannot be converted and should stop counting as pipeline. Right intent with the wrong next step is where the recoverable buyers are.
- Report buyers, not visitors. Split organic sessions by whether the page has a path to a conversion, and never delete or redirect a page without a query-level Search Console review.
Every organic traffic report has a few pages that look alive. They rank, they collect clicks every day in October and every other month, and nobody who lands on them ever buys, books or signs up. A zombie page is a page that earns organic visits and produces no conversions and no assisted conversions. You find them by setting organic sessions per landing page against both kinds of conversion, and you deal with them by reading the queries behind the visits before changing anything.
That is the end of the Halloween part. The rest is a method for a company that already invests in SEO, already has rankings, and wants to know which of that traffic has a buyer behind it. It covers how to find these pages in Search Console and GA4, where those reports stop being reliable, how to tell a page that cannot convert from one that could, the five decisions available, and what the exercise changes in how SEO ROI is reported.
It takes a different angle from our earlier piece on why a SERP rank is not generating sales, which looks at the visitor. This one looks at the page inventory: which URLs to work on, which to stop counting and which to leave exactly as they are.
What a zombie page is, and what it costs
The definition has two conditions, and the second is the one most audits skip. A page with no conversions in the session is a suspect. It becomes a zombie only when the people who landed on it also fail to convert on any later visit, which is what an assisted conversion would show.
The cost is rarely the page itself. A published article costs almost nothing to keep online. The cost is in the decisions made from the numbers it produces:
- The SEO report overstates progress. Sessions rise, the conversion event does not, and the sitewide organic conversion rate falls for reasons that have nothing to do with the pages that sell. The same blending problem is behind most cases of a conversion rate that looks low because of traffic quality.
- The content plan copies the wrong winners. If the brief for next quarter is built from the pages with the most traffic, it reproduces the pages with the least commercial value.
- Recoverable buyers go unnoticed. Some of these pages receive visitors who are deciding. They sit in the same list as the pages that never will, so nobody works on them.
A page with no conversions in the session is a suspect.
The stakes are higher when organic clicks are harder to earn. If you are already dealing with a website that is losing organic traffic, knowing which of the remaining visits could produce a buyer matters more than the total.
How to find zombie pages in Search Console and GA4
The audit is one table: organic landing pages as rows, then sessions, same-session conversions and later-visit conversions as columns. No report in Search Console or GA4 produces that table on its own, so it is assembled from three.
- Search Console, Performance report, Pages tab. Set the date range to at least three months, sort by clicks and export. Then click a page and switch to the Queries tab to see what it is found for. This gives you clicks, never outcomes.
- GA4, landing page view filtered to organic search. Use the Landing page report with a filter on session default channel group, or, if Search Console is linked, the Google organic search traffic report, which Google describes as showing landing pages with Search Console and Analytics metrics side by side. Read sessions and key events per page, one key event at a time.
- GA4, exploration with a user segment per page. For each suspect, build a segment of users whose session started on that page, then count the same key event across all of their sessions in the period. This is the assisted conversion check. The key event attribution paths report cannot do it for you: it shows early, mid and late touchpoints by channel group, source, medium or campaign, with no landing page dimension.
- Join on URL and filter. Keep pages above your minimum volume with zero in both conversion columns. That is the candidate list.
Where each report stops being reliable
| Report | Use it for | Limit to plan around |
|---|---|---|
| Search Console, Performance (Pages and Queries) | Organic clicks per page and the queries behind them | Some queries are withheld for privacy, and only the most important rows are stored and shown, so the query list for a page is partial |
| GA4, landing page view, organic only | Sessions and same-session key events per landing page | Credits only the session that started on the page. A purchase three days later appears against a different landing page |
| GA4, exploration with a user segment | Whether users who landed on a page converted on any later visit | Depends on GA4 recognising the returning user. Device changes and declined consent break the link, and small segments can be withheld by data thresholds |
| GA4, key event attribution paths | How often organic search assists as a channel | Channel level only. It cannot name the page that did the assisting |
Two of those limits come straight from Google. The Search Console documentation states that anonymized queries are omitted from the report to protect user privacy, and that the table stores only the most important rows. GA4 applies data thresholds to reports and explorations when user counts are low. Both mean the audit undercounts slightly, in the direction of making a page look more dead than it is.
The minimum volume before zero means anything
A page with 80 sessions and no conversions has told you nothing. The quick test is the rule of three: zero conversions becomes meaningful at about 3 divided by the conversion rate you would expect from that page. The calculation: at an expected rate of 1%, the chance of seeing zero conversions in 300 sessions by luck alone is 0.99 to the power of 300, about 5%.
Common mistake: building the zombie list from the GA4 landing page report alone. That report only sees the session that started on the page, so every top-of-funnel article that feeds a later purchase, demo or appointment lands on the list. Teams then prune or rewrite the pages that open their best journeys. Run the later-visit check before any page is labelled.
Wrong intent or wrong next step: telling them apart
Every page on the candidate list has the same symptom. The cause is one of two, and they call for opposite responses.
A page that cannot convert has the wrong intent. Its queries ask for a definition, a fact or a method with no purchase behind them. The visitor gets the answer and leaves satisfied. No button changes that, and adding a hard call to action to such a page mostly produces a worse page.
A page that could convert has the right intent and the wrong next step. Its queries compare, price or shortlist something you sell. The visitor is deciding. The page answers the question and then offers nothing relevant, or asks for the final commitment when the visitor wanted one more piece of evidence.
The test is done in Search Console at query level, page by page. Open the page, read its top queries by clicks, and mark each one as buying intent or not. Then look at what the page offers after the answer. The guide to combining CRO and SEO covers the intent mechanics in depth. Here the only output needed is a label per page.
- Queries are definitions, facts or how-tos
- No query names a product, price or provider
- Visitors rarely view a second page
- Queries compare, price or shortlist
- Visitors move on to product, pricing or booking pages
- The only call to action is generic or sits in the footer
- The keyword the page was written for
- Bounce rate on its own
The same audit for a store, a SaaS site and a clinic
What counts as the next step depends on the conversion event, and each business model has one. An online store counts purchases. A B2B SaaS site counts demos or trials. A service business such as a clinic counts appointments. Run the audit once per event and keep the results apart: a page can be a zombie for purchases and an earner for newsletter sign-ups, and adding the two would hide exactly what the audit is for.
The decision for each page
There are five decisions. Three of them involve no change to the page at all, which is the point: the audit is mainly a way of deciding where not to spend effort.
| Decision | When it applies | What changes | Risk to rankings |
|---|---|---|---|
| Keep as is | No same-session conversions, but its visitors convert later | Nothing on the page. It gets credited as an assist in reporting | None |
| Add a next step | Buying-intent queries, content answers them, nothing relevant follows | The offer, proof and call to action after the answer. The body copy stays | Low, the ranking content is untouched |
| Rewrite for intent | Buying-intent queries, but the page answers a different question | The content, rebuilt around the queries it is actually found for | Medium, review the queries first and keep the URL |
| Consolidate | Two or more pages serve the same queries and split the clicks | One primary page absorbs the others, which redirect to it | Medium to high without a per-URL query check |
| Leave alone | No buying intent, no overlap, no cost to keep | Nothing on the page. Its sessions stop counting toward SEO ROI | None |
Adding a next step is the decision with the best ratio of return to risk, because the page already ranks for the right queries and the content that earns the ranking does not change. The step has to match where the visitor is: a comparison reader wants the options with prices, and a first-time reader of a treatment page wants to know who performs it before being asked to book. This is the part Pathmonk automates. It reads in-session behaviour to estimate how close each visitor is to a decision and shows a matching microexperience on top of the page, so the indexed content stays as it is.
Consolidation is the decision with the most ways to go wrong. Before any page is merged or redirected, export its queries from Search Console and check three things: which queries it holds that the primary page does not, whether it has external links, and whether the primary page can absorb those queries without losing its own focus. No page is deleted or redirected without that query-level review. A page that looks dead in the conversion column can still be the only thing holding a set of rankings in place.
A worked example: six pages, 52,000 sessions
Hypothetical example
An online store selling coffee equipment audits its organic landing pages over 90 days. The conversion event is a purchase. The site expects roughly a 1% purchase rate from organic visits, so the minimum volume is 300 sessions per page. Six pages are shown here. All figures are invented to show the method.
| Landing page | Organic sessions | Purchases, same session | Purchases, later visits | What the queries show | Decision |
|---|---|---|---|---|---|
| Guide: descaling an espresso machine | 18,400 | 0 | 61 | Owners maintaining a machine | Keep as is |
| Explainer: what a flat white is | 12,600 | 0 | 0 | Definitions, no buying intent | Leave alone |
| Comparison: burr or blade grinder | 9,200 | 0 | 0 | Comparing before buying, page links to no products | Add a next step |
| Article: coffee to water ratio | 7,900 | 0 | 0 | Mostly people looking for a coffee scale | Rewrite for intent |
| Guide: French press | 2,100 | 0 | 0 | Same queries as the how-to below | Consolidate |
| How-to: using a French press | 1,800 | 0 | 0 | Same queries as the guide above | Consolidate |
On same-session data alone, all six pages are zombies and the report says 52,000 sessions produced nothing. The later-visit check changes that. The descaling guide has 61 purchases from people who first arrived on it, so it is assisting and stays as it is. The other five have nothing in either column: 12,600 + 9,200 + 7,900 + 2,100 + 1,800 = 33,600 sessions, which is 65% of the 52,000.
The work that follows is small. One page gets a next step, one gets rewritten around the queries it is found for, two are merged after their queries are compared line by line, and one is left alone and removed from the pipeline numbers. Nothing is deleted.
How this changes SEO ROI reporting: buyers, not visitors
Once pages are labelled, the SEO report can stop treating every session as equal. Three lines replace the single traffic number:
- Organic sessions, split by page type. Sessions on pages with a path to the conversion event, and sessions on pages without one. Growth in the second group is reported, and is not presented as progress toward revenue.
- Buyers with an organic landing in their journey. Same-session and later-visit conversions, for one event. Purchases for a store, demos or trials for a SaaS site, appointments for a clinic.
- SEO cost per buyer. SEO spend for the period divided by line 2. This is the figure that can sit next to paid acquisition cost for the same event.
Our own published SEO results show why the distinction matters. The Pathmonk SEO case study reports an average organic traffic lift of +35% and an average organic impressions lift of +63% across Pathmonk SEO customers, comparing the three months after install against the three months before it. Those are visitor-side numbers. The line in that study a finance lead would read first is a different one: Joto Imaging Supplies, a B2B e-commerce company, with $60k extra organic revenue at month 3. Traffic growth is the input. The buyer-side figure is the result, and a report should carry both.
If the audit shows that most of your organic growth is landing on pages with a buyer path, the traffic number was telling the truth. If it shows the opposite, the next SEO investment is in next steps and intent, and the SEO work itself should be briefed from the pages that produce buyers.
- Organic sessions
- Rankings and impressions
- Sitewide organic conversion rate
- Organic sessions on pages with a buyer path, and without
- Buyers with an organic landing in their journey, one conversion event
- SEO cost per buyer
Top-of-funnel pages that feed later visits are assisting, and the method exists to protect them. Low-volume sites and low-volume pages do not have enough sessions for zero to mean anything: below the minimum in the rule of three, extend the date range or wait. Long sales cycles need a window at least as long as the cycle, or later conversions are missed. And where returning visitors are hard to recognise, because of device changes or declined consent, treat a clean zero in the later-visit column as probable, never certain.
Key takeaways
- A zombie page earns organic visits and produces no conversions and no assisted conversions. Both columns have to be empty.
- No single report shows it. Join Search Console clicks and queries, GA4 same-session key events and a GA4 exploration for later-visit conversions on the page URL.
- Zero conversions only means something above a minimum volume: about 3 divided by the expected conversion rate, in sessions.
- The queries decide the diagnosis. Wrong intent cannot be converted. Right intent with the wrong next step can.
- There are five decisions: keep as is, add a next step, rewrite for intent, consolidate, leave alone. Deleting is not one of them.
- Nothing is consolidated or redirected without a query-level Search Console review of every URL involved.
- Count one conversion event per audit. Purchases, demos or trials, and appointments stay in separate columns.
- Report organic sessions split by whether the page has a path to a buyer, and SEO cost per buyer, alongside traffic.
FAQs on zombie pages and organic conversions
What is a zombie page in SEO?
In this article, a zombie page is a page that earns organic visits but produces no conversions and no assisted conversions. It is alive in the traffic report and absent from the revenue report. The term is also used elsewhere for pages with no traffic at all, which is a different problem.
How do I find organic pages that get traffic but do not convert?
Join three reports on the landing page URL. Use the Search Console Performance report for organic clicks and queries per page, the GA4 landing page view filtered to organic search for sessions and same-session key events, and a GA4 exploration with a user segment per page to check for conversions on later visits. Pages with real traffic and zero in both conversion columns are the candidates.
Should I delete zombie pages?
Rarely, and never as a first step. Most zombie pages should be left alone, given a next step, rewritten for the intent of their queries or consolidated with an overlapping page. No page should be deleted or redirected without a query-level Search Console review, because a page with no conversions can still hold rankings, links and queries that another page depends on.
How much traffic does a page need before zero conversions means something?
Use the rule of three: divide 3 by the conversion rate you would expect from that page. At an expected rate of 1%, a page needs about 300 organic sessions before zero conversions is unlikely to be chance. At 0.5% it needs about 600, and at 0.2% about 1,500. Below that, extend the date range instead of judging the page.
Do informational pages count as zombie pages?
Only if the people who land on them never convert later either. A top-of-funnel page with no same-session conversions that starts journeys which close on a later visit is assisting, and it should be kept and credited. A definitional page whose visitors never return and never convert is a zombie, and the right decision is usually to leave it alone and stop counting its visits as pipeline.
How should SEO ROI be reported if not by traffic?
Report the conversion event the business runs on, one event at a time: purchases for an online store, demos or trials for a B2B SaaS site, appointments for a clinic. Show organic sessions split by whether the landing page has a path to that event, the number of buyers whose journey included an organic landing, and the SEO cost per buyer for the period.
- Google Search Console Help: Performance report, dimensions and data groupings (support.google.com, accessed 5 Oct 2026)
- Google Analytics Help: Google organic search traffic report (support.google.com, accessed 5 Oct 2026)
- Google Analytics Help: key event attribution paths report (support.google.com, accessed 5 Oct 2026)
- Google Analytics Help: about data thresholds (support.google.com, accessed 5 Oct 2026)
- Pathmonk: SEO results case study (pathmonk.com, figures as published on 5 Oct 2026)