I pulled ninety days of Google Search Console data this week and sorted by impressions descending. Then I filtered for pages sitting at position eight or nine — the bottom of page one, just inside the visible window. Then I sorted those by click-through rate, ascending.
The first page on the list had been shown nine hundred and fourteen times in ninety days. It had earned one click.
I’d written the page. I’d published it. I was happy with it. And for ninety days, when somebody searched for what it covered and Google decided to show them my result, almost nobody clicked.
The page wasn’t the problem. The title was.
What the Title Was Doing
The article in question was about product photography — specifically, the gap between how a customer assesses a product in a shop versus how they have to assess it on a screen. Your customers can’t pick it up, turn it over, or feel the weight. That’s the whole argument in one line.
The title I’d given it was “Boosting E-commerce Sales with Product Images: Building Trust Through What You Show.”
Eighty-three characters. By the time Google rendered it in the results page, with the brand suffix and the truncation rules, what a searcher actually saw was something like “Boosting E-commerce Sales with Product Images: Building Trust Through Wha…”
Two abstract verbs at the front. The promise — what the reader actually gets — buried at the back. And the back got cut off before anyone read it.
A title that gets truncated isn’t a title. It’s a sentence Google chose to stop reading on your behalf.
I rewrote it as “Product Images That Build Trust and Sell.” Forty characters. Renders cleanly with the brand suffix at fifty-eight. The claim is in the visible window. The promise is the thing the searcher gets on the click.
That was the first one.
The Four Rewrites
I did the same exercise three more times.
Pain Point Matrix had been getting eleven hundred and forty-five impressions and three clicks. Old title: sixty-nine characters, framework name plus a value-prop philosophy that got chopped. New title: “The Pain Point Matrix: Find What to Write.” Forty-one characters. The framework name is the query people type — keep that. The rest just has to tell them what they walk away with.
UK Shopify Shipping had been getting fifteen hundred and fifty-three impressions, nineteen clicks, and a one-point-two-two percent click-through rate. The title started with “UK Shopify Shipping Rates” — but the search query people actually type is “shopify shipping rates uk.” So I reordered the lead phrase to match. Then I dropped the “Stops You Losing Money” tail that was eating characters without earning the click. The new title is “Shopify Shipping Rates UK: Postcode Fix.” Thirty-nine characters.
EKM to Shopify Migration was sitting at one-point-eight-six percent — better than the others, but still well below where it should be at position five. Same treatment. Tighter title. Visible window protected.
Four pages. Four titles. None of them touched the body of the article.
Why This Matters
Most of what I’ve written about marketing this year has been the gap between what looks like it’s working and what actually is. A few weeks ago it was a single line of code telling Google every page on a site lived at a domain that didn’t exist — site looked fine, traffic was bleeding out invisibly. Before that it was a Google Ads account double-counting every conversion, making the cost-per-lead look half what it really was. The same shape sits beneath why a site can get visitors but no enquiries — the layer you don’t inspect is the layer doing the damage.
This is the same story, viewed from a different angle. A page can be ranking. It can be in front of the people you want it in front of. And it can be earning almost nothing, because of a layer most people never inspect — the line of text Google actually shows in the search results.
A page that gets nine hundred impressions a month and earns one click isn’t a failing page. It’s a working page being failed by the bit that introduces it.
The check in one line
Open Google Search Console. Performance report. Last ninety days. Filter to position one through ten. Sort by click-through rate ascending. The first three pages on that list are almost certainly being failed by their titles, not their content.
What to Look For
If you want to do this yourself, the signal is simple. A page is sitting on page one of Google. It’s earning thousands of impressions over a ninety-day window. The click-through rate is under two percent.
That’s the shape. Lots of people are seeing it. Almost none of them are clicking.
The cause, ninety percent of the time, is one of three things. Either the title is too long and the meaningful part is getting truncated. Or it leads with abstract framing — Boosting, Optimising, Mastering — instead of a concrete promise. Or the order of the words doesn’t match the order people type the search.
The fix, in every case, is to write a title that is shorter than fifty-five characters, leads with the noun the searcher typed, and ends with a claim — what the reader actually gets on the click. The whole exercise takes about twenty minutes per page once you’ve got the data in front of you.
I’ll re-pull the same Search Console data fourteen days from now and see what changed. The rule of the room is that titles affect click-through rate within days, not months — Google reads the new title within a crawl or two and starts serving it. If the rewrites worked, the click-through rates on those four pages will be visibly higher. If they didn’t, I’ll know something else is wrong.
That’s the cycle. Pull the data. Find the page that’s being failed by its own introduction. Rewrite the introduction. Re-measure. The page itself doesn’t change. The number of people clicking on it does.
Tony Cooper
We Build Stores
tony.cooper@webuildstores.co.uk
07963 242210
P.S. If you’d like me to run the same exercise on your site — pull the ninety-day data, find the under-performing titles, send you a list with rewrite suggestions — reply to this email and I’ll do it. No charge. The pages that are sitting on page one and earning nothing are the easiest wins on most sites, and they’re the ones nobody looks at, because by every other measure they look fine.
