Lifting a coffee store from just below the first page. 110% more keywords on page one.
A specialty roaster whose online range runs from single-origin beans to flavoured, decaf and low-acid blends was showing up a lot, but mostly just below the fold of page one. I worked on its product pages first, then its content and collection structure. Comparing the first three months of the project window with the latest three, the store ranked on page one for 110% more searches, and its average position crossed into the top ten.
Results
Keywords on page one
Keywords in the top three
Plenty of products, too few of them on page one
In October 2025 the store's average position was about 13. That is page two, where shoppers rarely scroll. Of all the searches that surfaced the site that month, only about a third reached the first page, and fewer than a hundred the top three.
The catalogue was the bigger issue. More than a hundred products sat across nearly 300 URLs, and each needed a title, heading and description written for the terms people use to find that exact coffee, so pages would not compete with each other. There were around a dozen collections, and not one grouped coffees by where they were grown, even though origin searches carry obvious buying intent. And the roaster had a real edge in low-acid coffee that the site was not yet using to answer pre-purchase questions.
Product pages, then guides, then shelves
My project summary records what was delivered, not the date each piece went live, so the order below is the order of priority rather than a timeline.
- 01
Give every product one search to own
I built a metadata workbook for every product URL in the store, with short, keyword-led titles and no two titles or descriptions alike. That stops pages cannibalising each other. I then rewrote the heading, title and description directly in the store for roughly thirty products, about half of them single-origin coffees and the rest flavoured coffees, blends and decafs.
- 02
Tidy the URL set
An audit of the store's addresses flagged duplicate and misspelled product URLs for redirection, and marked utility pages to be kept out of the index. The aim was simple: one clean address per product, so ranking signals gather in one place instead of splitting across near-copies.
- 03
Plan content around the low-acid edge
I researched about sixty informational keywords and grouped them into ten topic clusters, checking each one against every existing URL to avoid overlap. Low-acid coffee went first. Four posts on it are published, from what the term means to whether darker roasts really carry less acid, and four more on roasting and origin are being written to link back into them and the products.
- 04
Map the catalogue into collections
Every product was assigned to one of about two dozen proposed collections, nearly half of them by country of origin. That gives origin searches a dedicated landing page instead of leaving them to individual product pages or nothing at all.
Into the top ten, on average
Across three-month windows, the store ranked on page one for about 1,300 searches at the end against roughly 640 at the start, a rise of about 110%. Top-three rankings grew 80%, and the total number of searches showing the site rose 100%. Impressions per window rose by a little over 50%.
Month by month, September 2026 held about 130% more page-one keywords than October, and average position moved from about 13 to 10. The four low-acid guides went from nothing to roughly 5,700 impressions in their first three months, with two already ranking on page one. The gains followed the work.
Other case studies
Want a result like this for your site?
A 30-minute call, no pitch deck. Tell me where you are stuck and I will tell you honestly whether I can help - and if I cannot, who can.