A ranking can cost money when the people it brings were never going to buy and the page still takes time to defend. Below is the table that exposes it, from our own Search Console for 31 August to 24 September 2026, and the method for building yours: traffic, intent and conversions side by side, by landing page.
How a ranking costs money
Ranking is not conversion, and holding a ranking uses real resources whatever it converts at. A broad, informational search can bring thousands of visits a month from people who are researching, studying or solving a neighbouring problem your product does not address.
The cost appears in two places most monthly reports never show. First, the content refreshes and technical attention spent holding the position, because competitors keep publishing. Second, the opportunity cost: the same hours could have gone to a smaller search that sits much closer to a purchase.
Our own site, read page by page
Before our site moved to its new build at the end of September 2026, we reviewed which pages earned impressions and clicks. The table is from Google Search Console for our own site, read through Ahrefs, for 31 August to 24 September 2026.
Pages are described rather than named. Over those 25 days our home page, mostly brand searches, took 130 clicks.
- Blog listicle of app ideas1,786
- Home page (brand searches)1,326
- eCommerce marketing service page39,628
- Google Ads service page11,740
- Blog roundup of keyword research tools11,586
- Blog listicle of app ideas47
- Home page (brand searches)130
- eCommerce marketing service page2
- Google Ads service page17
- Blog roundup of keyword research tools1
| Page | Impressions | Clicks | Average position |
|---|---|---|---|
| Blog listicle of app ideas | 1,786 | 47 | 21.1 |
| Home page (brand searches) | 1,326 | 130 | 12.6 |
| eCommerce marketing service page | 39,628 | 2 | 33.6 |
| Google Ads service page | 11,740 | 17 | 32 |
| Blog roundup of keyword research tools | 11,586 | 1 | 37.1 |
Two patterns sit in five rows. The service pages had tens of thousands of impressions and almost no clicks, because they ranked on page three or four. That is a visibility problem, not the one this post is about.
The first row is. The app ideas listicle was the page with the most non-brand clicks on our site in that period, and its searches included Indonesian-language queries for app ideas. Those readers wanted ideas, not a development partner. On a traffic report it looks like our best blog post; on intent, it attracts almost nobody we can serve.
Our own triage had already marked it as off-subject for retirement, and the review flagged that retiring it would give up those clicks. That is the trade this post describes: visits that will not become clients, against the cost of keeping a page alive for them.
Check the ranking report against Search Console first
The same review found a second trap. In late September a rank tracker estimated a new page-one ranking worth 157 visits for one search, “ai development companies”. For 31 August to 24 September, Search Console showed 37 impressions, an average position of 18.3 and no clicks for it.
Estimated traffic is a model; Search Console records what Google actually showed and what people clicked. Before you decide a keyword is valuable, or that it is costing you, read the real numbers.
Build the table for your own site
You need Search Console and an analytics tool with conversions by landing page. Search Console has no conversion column, so the two have to be joined.
- Export Search Console’s pages report for a settled period: impressions, clicks and average position per page.
- Export conversions by landing page for the same period from your analytics: sales, leads or sign-ups, whichever the business counts.
- Join them on the page address and add conversion rate per page.
- For the pages with the most clicks, read the top queries in Search Console and label each by intent: researching, comparing, or ready to buy.
- Mark any page whose clicks are high, whose conversion rate is far below your commercial pages, and whose top queries are mostly research.
Compare each page with your own commercial pages, not an industry average. Research pages will always convert lower than product pages; the question is whether the visits times the conversion rate still justify the hours spent holding the position.
What to do with a page you have marked
| What the table shows | Usual move |
|---|---|
| Research intent, no link to what you sell | Step down upkeep gradually; retire with a redirect only if the clicks are worth less than the page costs |
| Research and buying intent mixed on one page | Split into two pages, each written for one intent |
| Buying intent, low conversion | Not a keyword problem: fix the page through conversion work |
| A smaller, buying search with no page | Move the freed hours there |
The shape of the trade: “best running shoes for flat feet” is a comparison search, while “buy” plus a specific model is a ready-to-buy one. The second carries far less volume and a far higher chance of a sale. Planning which pages to build for which intent is content strategy work; the volume was never the issue.
When a low-converting ranking is worth keeping
Do not cut every page that converts poorly. Keep it when:
- It earns links or citations that lift the pages that do sell.
- It is the first contact in a long buying cycle, and your analytics show those readers returning later through other pages.
- Upkeep is close to nothing, so holding it costs little.
And step down gently. Stopping all upkeep at once can let the ranking decay fast enough to lose what residual value it carries. The same logic applies to paid search, where more spend can stop paying for the same reason.
Written by
Shweta Agarwal
Chief Operating Officer, Vinove
Shweta runs the operating side of the group, holding delivery, process and accountability consistent across the group. Her focus is on the systems that let teams grow without losing the standard they started with.
LinkedIn →
