Case study · Shopify · Automotive diagnostics
From page four of Google to 3.4× the clicks
An online shop selling dealer-level diagnostic software for European car brands already appeared for the right searches, on pages four to six. We worked on it for six weeks over July and August 2026. In the 38 days after the work ended, Google sent it 653 clicks. In the 38 days before we started, it sent 194.
- Platform
Shopify
- Catalogue
- About 260 products
- Languages
- English and Spanish
- Engagement
- 25 Jul to early Sep 2026

- Organic clicks
- 653was 1943.4×
- Times shown
- 25.4Kwas 16.0K+59%
- Click rate
- 2.6%was 1.2%+1.4 points
- Avg. position
- 22.1was 38.416 places up
Clicks per week, June to August
work starts
Before the work the shop got 27 to 41 clicks a week. The five full weeks after we started brought 66, 79, 83, 87 and 114.
The "after" window starts when the engagement ended, so these are gains the shop kept with nobody working on it. Position 38 is page four of the results; position 22 is the top of page three.
Why page four was the problem
Google already showed the shop to the right people. Workshops typed a car brand plus "dealer diagnostic tool", or the name of the factory software, and the shop came up. It came up on page four, five or six: 81% of its impressions sat below position 30.
81%of its impressions sat below position 30, on pages four to six
How many saw it per click · share of impressions
- Page onepositions 1 to 101 in 183%
- Page two11 to 201 in 899%
- Page three21 to 301 in 1066%
- Pages four and five31 to 501 in 53143%
- Page six and beyond51 and lower1 in 1,05339%
Bar length is click rate, scaled to page one. Search Console query data, 1 Jun to 31 Aug 2026, covering the 27,307 impressions that came with a reported search term.
What workshops typed · avg. position, Jun to Aug
- bmw dealer diagnostic tool47.1
- mercedes dealer diagnostic tool45.1
- vw audi skoda seat dealer diagnostic tool54.2
- bmw ista software32.2
- jlr sdd coding programming tool61.6

- 1 in 18
- searchers clicked on page one
- 1 in 1,053
- clicked below position 50
The pages were the same in both cases. Moving them up the results was the job.
What we did
We started with a 65-item checklist across 12 areas and a scored audit that we re-ran three times as fixes landed. The overall score went from 63 to 71 in the first session.

- 0134 to 94
Guides written by car manufacturer
A 62-article plan with one search term and one intent per page, then guides on the factory software workshops search for: BMW ISTA and ICOM, VW group ODIS and VAS6154, Mercedes Xentry, Porsche PIWIS, Jaguar Land Rover SDD, plus buying and comparison pieces. Guides became the best-ranking page type on the site, at an average position of 28.9.
- 0245 to 82
Structured data Google can read
The site published nothing beyond Shopify's defaults. We added shop and website data with site search on every page, full product data with SKU, shipping, seller and condition, and breadcrumbs on product, collection, blog and page templates. Thirteen automated checks render the templates before each push.
- 03417,539
Characters of dead markup removed
API access to the shop was blocked, so we cleaned all 287 products from the catalogue export with a scripted pipeline. It removed 1,600+ emoji, pasted styling and tracking residue, and 56 duplicate H1 headings, and wrote 129 missing Google descriptions (18 by hand, 111 from each product's own copy). Prices, SKUs, handles and all 418 images were checked byte for byte against the original before import.
- 040 fake
Real reviews in place of invented ones
The homepage showed ratings like "4.9 from 312 reviews" with no review app installed. We removed them, installed one, switched the rating display on, and caught a hard-coded five-star line in the widget copy before it went live. Product data only publishes a rating to Google when real reviews exist.
- 05EN + ES
Menus by car brand, Spanish rebuilt
The shop menu was reorganised by manufacturer, the legal and contact pages were rebuilt, and the whole catalogue was re-translated into Spanish from the cleaned text.
- 06284
Spam backlinks disavowed
Links from link-selling domains were identified and submitted to Google to be ignored.
Rankings for the searches that sell
The gains came from moving up on existing searches, which is where the extra clicks were.
Average position in Google Search Console, same 38-day windows as above. Searches are grouped and named generally to keep the client anonymous.
- Mercedes diagnostic software9.4 up, now page one
Before 15.7After 6.3
- VW group workshop software6.3 up, now page one
Before 13.4After 7.1
- Land Rover and Jaguar diagnostic software23.6 up
Before 57.1After 33.5
- Average, all searches16.3 up
Before 38.4After 22.1
Spanish was the strongest side of the shop
- Spanish pages
- 22.4avg. position · 2.81% click rate
- English pages
- 39.9avg. position · 1.15% click rate
1 Jun to 31 Aug 2026. Spain had the best click rate of any large market, at 3.52%. Of five competing shops we checked on 3 Sep, one published in a second language.
What the numbers don't prove
- 01This is a before and after comparison with no control group. Market demand and the time of year can move the same numbers.
- 02We have no reliable sales figures. The shop fired two Google Analytics properties, each more than once, so sessions and conversions were double counted. We flagged it as a critical fix in the shop's admin, and we don't report revenue for this engagement.
- 03Performance was the one audit category that went backwards in the first session, from 50 to 45.
- 04The shop is still mostly on pages two and three for its main terms, so the biggest gains are still available.
Figures from Google Search Console, measured to 11 October 2026. Workshop photos from Unsplash.
See our SEO packages
Ready to Build a Digital Business Around What You Do Best?
Tell us about your idea or project and let's create a digital setup that supports your goals, not the other way around.