A single location storage business whose website, in Google’s eyes, was really one page. Twelve months later its clicks from Google had more than quadrupled, and most of them were landing on pages other than the home page.
more clicks from Google in the last 28 days than the same 28 days a year earlier.
The business had a website, a service area and real demand. What it did not have was a site Google could do much with. Across a sixteen week window a year ago, only 33 pages appeared in Google at all, and just 18 earned a single click. The home page took 85% of every click the site received.
That is a fragile place to be. The whole pipeline rested on one page holding one ranking, and the average position across everything the site showed up for was 32, the fourth page of results.
The reports also made the site look bigger than it was. They listed 622 web addresses, but only 33 were real pages. The rest were the same few pages reached through tracking tags, page anchors and a second spelling of the domain, each counted separately and none earning anything.
About three new pages a week, every week, through the same four steps. The pace matters less than the discipline.
Every topic traces back to something measurable: searches the site already showed up for and lost, questions people asked alongside them, and gaps in what competitors had published. If nobody searches for it, it does not get written.
Each page starts as a short brief: what ranks now and why, what every competing result covers, what none of them cover, what the page has to prove, and which existing pages it should link to.
Twenty four automated checks run on each draft before it goes anywhere near the site, covering structure, readability, sourced claims and clear next steps for the reader. A draft that fails gets rewritten or held.
Approved pages go live spread across the week instead of in a batch, and each one is submitted to Google the moment it publishes. Then its results are measured against the brief that produced it.
Several pages targeted the same core service term the home page already ranked for, splitting one ranking four ways. Those pages were refocused on different search intent so the home page could hold the term alone.
Every tracking tag, anchor and alternate domain version of a page now resolves to one address. Nothing was deleted. The same content simply stopped competing with copies of itself.
Page titles and descriptions were written page by page to match what each page actually delivers, replacing generated text that Google was cutting off mid sentence.
Partway through, the data pointed at an easy win. A group of related searches was producing thousands of appearances a month, so the natural call was to write more pages there.
Checking it first killed the idea. Those were national searches, and on national searches the site averaged position 24.8. On searches tied to its own service area it averaged 7.7. More pages there would have bought appearances and almost no clicks.
The real opportunity was seven pages already on page one that nobody clicked. Three were competing with the home page. The rest ranked for searches their listings did not answer. Fixing them took a fraction of the effort of a new content cluster, and it is where the jump in clicks came from.
average position
average position
Seven pages were already on page one of Google with 5,356 appearances between them, and nobody was clicking.
That was the winnable problem, and the answer was fixing pages, not writing more.
Clicks from Google by month, June 2025 through September 2026. June 2025 is a partial month, drawn hollow.
| Quarter | Clicks | Appearances | Avg position |
|---|---|---|---|
| Four quarters ago | 451 | 49,347 | 36.6 |
| Three quarters ago | 520 | 64,995 | 32.3 |
| Two quarters ago | 605 | 64,987 | 21.8 |
| Latest quarter | 1,176 | 112,793 | 17.6 |
Both columns cover the same sixteen weeks, a year apart.
| What it measures | Before | After | Change |
|---|---|---|---|
| Pages appearing in Google | 33 | 212 | 6.4× as many |
| Pages with 100+ appearances | 25 | 132 | 5.3× as many |
| Pages earning at least one click | 18 | 117 | 6.5× as many |
| Different searches the site appeared for | 887 | 2,608 | 2.9× as many |
| Searches ranking on page one | 282 | 1,259 | 4.5× as many |
| Searches ranking in the top three | 171 | 685 | 4.0× as many |
This is the result that matters most, and it never shows up in a traffic total. The home page did not get weaker. It grew from 445 clicks to 564. Everything around it grew much faster, from 76 clicks to 803.
So the home page’s share of all clicks fell from 85% to 41%. If one ranking slips tomorrow, the business keeps most of its search traffic.
One number has not moved yet: click rate. It was 1.13% a year ago and is 1.09% now, while appearances more than quadrupled. All of the growth so far came from being found more often, not from more searchers choosing the listing.
An average position of 13.3 shows where the remaining value is. A large share of the site’s visibility now sits just below the top of page one. Pushing those rankings up the last few places, and rewriting the listings shown against them, should turn visibility the site already earned into more enquiries, without adding a single page.
Page count is not the goal. Clicks from searches the business can actually win are.
All figures come from the client’s own Google Search Console. Year over year comparisons use identical calendar windows a year apart. The business, its market, its pages and its search terms are withheld at the client’s request; the performance numbers are unaltered.
