Spares for an orphaned vehicle line, found by the people who need them. +28% summer clicks, year on year.
This store keeps electric utility vehicles running after their maker closed its doors. Demand rises and falls with the seasons, so I judge it by comparing each month with its counterpart twelve months earlier. I took on the SEO at the end of November 2025. Over the summer of 2026, Google clicks ran 28% ahead of the summer before, and the site sat in the top three for 450% more searches.
Results
Google clicks
One product line, no manufacturer, no blog
The business sells spares, plus kits that convert the vehicles to lithium power, for one family of electric utility vehicles. The company that built them is gone, so owners hunting for a part have nowhere official to go. That makes the store useful, but only if search can connect a model and a broken component to the right product page.
When I came in, the site had no blog, and the client wanted one. Across the first three months of keyword data it appeared for roughly 400 searches, and only a few dozen of those put it in the top three. The client was also clear that trade peaks in the autumn and slows in winter, which is why every comparison here pairs a month with its twin from the year before.
Two jobs: a catalogue buyers can browse, then answers to what broke
My notes for this account record scope and outputs, not a dated diary, so I describe what was planned and shipped without pinning each item to a week.
- 01
Read the whole inventory before proposing anything
In the onboarding answers, the client singled out the lithium kits and the braking components as the lines that matter most. Before suggesting a single new collection, I went through the full product inventory so every proposal would map to products that actually existed, rather than to keyword ideas with nothing behind them.
- 02
Build collections in four layers
I designed a collection plan on four levels: the original maker, each vehicle model, each type of part, and model combined with part type. It came to roughly 210 collections, about 180 of them new, each tied to real product URLs. The rollout works through that backlog in stages rather than all at once.
- 03
Write for the fault, not the part number
Owners often search for a symptom before they know which part has failed. I scored a dozen or more noise and fault searches into three priority bands, starting with the ones that point to a single part. Each article follows the same order: what the owner notices, what usually causes it, a way to check, and finally a link to the exact part. A handful were published by the time of the July 2026 report.
- 04
Report on clicks first, across every engine
Each monthly report looks at Google, then Bing, then citations in AI answers, and treats clicks as the number that counts.
Each summer month ahead of its 2025 twin
June 2026 finished about 24% above June 2025, July about 36% above, and August about 25% above. Taken as one block, the summer was up 28%. August 2026 brought roughly 1,500 clicks, against about 1,200 a year earlier. Clicks also climbed every month from January to August 2026, though some of that rise is simply the season turning, which is why I lean on the year-on-year view.
The footprint widened at the same time. Comparing the first three months of keyword data with the latest three, ranking keywords grew about 120%, page-one keywords about 200%, and top-three keywords 450%. Collection pages took roughly 57% more clicks in the three months to the end of July 2026 than in the three before. I read that as timing that fits the work, not proof that the work caused it. On Bing, Copilot cited the site about 170% more often in July than in February.
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.