241,000 Impressions, 1,042 Clicks: What Fact-Lookup Pages Earn in 2026

One of our sites is a pure data site: a page per UK council, each listing council tax bands and charges. In the 48 days from 11 August to 27 September it earned 241,047 impressions in Google Search and 1,042 clicks. That is one click for every 231 times the page was shown, at an average position that is often between 6 and 9.

Across seven sites, a pattern shows up that anyone publishing facts, data or calculators should see before they plan their traffic. It also contains one result about freshness that surprised us.

Impressions per click, across seven sites

Switch between click rate and impressions per click, and tap a site to see which of its pages earn the clicks.

Seven sites, one patternInteractive

How many impressions does it take to earn one click?

Switch the measure, then tap a site to see which of its pages earn the clicks.

Impressions per clickClick rate

• Padel Chippenham3 months to 27 Sep2.8
• georanklocal.com28 days to 27 Sep48
• sleepertrain.co.uk3 months to 27 Sep58
• File-conversion tools site6 months to 5 Aug215
• Council-tax data site11 Aug–27 Sep231
• Accident-statistics site28 days to 26 Aug389
• Tax-code reference site28 days to 26 Aug452

Council-tax data site

1,042 clicks from 241,047 impressions · Fact lookup, one page per council

Page or query	Clicks	Impr.	CTR	Pos.
Council page 1	33	2,233	1.5%	8.24
Council page 2	24	2,013	1.2%	6.65
Council page 3	18	978	1.8%	7.14
Council page 4	16	1,476	1.1%	9.03
Council page 5	15	1,207	1.2%	7.97

Answer / lookupBuying intentBrand / local

The sites sort themselves by what the page does for the searcher.

• A local venue, searched by name needs 2.8 impressions per click. People looking for Padel Chippenham click Padel Chippenham.
• A marketplace page for buying a railway carriage needs about 7 to 9. People ready to buy leave Google to do it.
• Fact and lookup pages need hundreds. The council-tax pages sit at 231, a tax-code reference site at 452 (28 days to 26 August) and an accident-statistics site at 389 in the same window.

Look at the pages themselves, not the site averages. The council-tax pages sit at positions 6.7 to 9.0 and earn between 1.1% and 1.8%. Page one of Google, bottom half, and roughly one click in 70. The sleeper-train marketplace's buying pages sit at similar positions (6.6 to 10.4) and earn 10.8% and 13.5%.

Why, as far as we can tell

We can't see inside Google's results for these queries, and we have not exported the Search Console AI features report for these sites, so this is a hypothesis: a searcher asking "how much is band D council tax in Preston" wants one number. If Google, or an AI summary, can show it without a click, a click isn't needed.

Pew Research's 2025 study of 900 US adults found 8% of visits clicked a traditional result when an AI summary appeared, against 15% when it didn't. Our numbers are not Pew's, but they point the same way: the more a query can be answered in one line, the fewer clicks it sends.

The one thing that changed the click rate: the year

Here is the part we did not expect. Look at the council-tax site's top queries and nearly all of them contain the tax year: "south holland council tax bands 26 27", "east devon council tax bands 2026/27", "wealden council tax bands 2026".

Freshness in the dataInteractive

Searchers add the year. Those searches click.

From the council-tax data site: queries that name the current tax year versus everything else.

Click rateShare of impressions vs clicks

Searches that include the tax year

2.01%

All other searches

0.42%

Year-stamped searches clicked about five times as often (2.01% against 0.42%).

Top year-stamped queries	Clicks	Impr.	Pos.
“south holland council tax bands 26 27”	5	34	4.91
“conwy council tax bands 26 27”	4	10	3
“east devon council tax bands 2026/27”	3	59	6.83
“wychavon council tax bands 26 27”	3	28	7.82
“wealden council tax bands 2026”	2	48	7.29

Among the 1,000 queries Google lists for the site, 95 name the tax year. They are 9.1% of the listed impressions but 32.1% of the listed clicks, a click rate of 2.01% against 0.42% for everything else. About five times higher.

We want to be careful: Google lists only about 10% of this site's impressions by query, positions differ slightly, and a few of these queries are very specific to one council. But the direction is hard to miss. People who add the current year are checking whether the page is up to date, and a page that is clearly marked as current gets the click.

That is freshness doing real work, and not by tricking a date field. The pages carry the 2026/27 charges, label them as such, and are the only version of that fact the searcher needs to open.

One place the pattern broke

An accident-statistics site in a crowded claims niche is an interesting exception. Its head terms ("motorbike accident compensation", "pedestrian injury claim") sit at positions 37 to 59 and earned zero clicks from 740 and 503 impressions. But two city statistics pages, one for Birmingham and one for Liverpool, sat at position 7 and earned 14.6% and 9.1% click rates.

The samples are tiny (41 and 33 impressions) and we would not build a strategy on them. But they match a plausible idea: a page with local, specific numbers that a summary can't compress is one a searcher has to open.

What to do about it

1. Treat lookup pages as visibility, not traffic. Don't forecast clicks from impressions. Measure impressions per click for each page type and plan from that.
2. Give the click a reason. Something a summary can't hold: a downloadable table, a calculator, a comparison across areas, a map, a worked example, a "what this means for you" section.
3. Put the year in, and mean it. If your data changes every year, say which year in the title and the first line, and update the page when the data changes. Do not change the date on its own.
4. Build the page that follows the lookup. Searchers who check a fact often have a next question: what it costs, who can do it, what to do about it. That page is where the click lives.
5. Check the AI features report. Search Console has a report for how your pages appear in Google's AI features. We pulled it for our own site, but not for these. It is the first place to look.

Method and limits

• Source: Search Console exports for seven sites. Windows differ (shown in the chart), so compare the pattern, not exact figures.
• "Impressions per click" is impressions divided by clicks over each window.
• The year-stamp analysis uses the top 1,000 queries Google lists for the council-tax site, which cover about 10% of its impressions. Year-stamped means the query contains "2025", "2026", "26 27", "26/27" or similar.
• Four of the seven sites are described by type, not name. The two named sites are our own case-study sites.

Sources

• Pew Research on AI summaries and clicks, summarised by ODSC

Read next: Same Google position, ten times the clicks