Back in February, a reader sent us a screenshot that we couldn't stop thinking about. Her Shopify store — a mid-sized apparel and accessory shop with roughly 1,400 SKUs — had lost about 41% of its organic sessions over four months. Google Search Console showed impressions holding steady. Clicks had collapsed. She'd already paid two freelancers, run a site audit through a well-known crawler, and come away with a 90-page PDF full of "missing meta descriptions" and "images over 100 KB." None of it explained the drop. We asked if we could follow the recovery as an editorial case study, with her shop kept anonymous. She agreed, and what unfolded over the next eleven weeks is the most instructive e-commerce SEO story we've published this year.
Week 1–2: The Audit That Missed the Actual Problem
The shop's first instinct — understandable — was to assume a penalty. There wasn't one. No manual action in Search Console, no ranking wipeout on branded terms. The bleeding was concentrated in non-branded category queries: "linen wide-leg pants," "everyday leather tote," that kind of long-tail commercial intent. Rankings hadn't vanished; they'd drifted from positions 3–6 down to 11–18, which is the quiet death zone where click-through rates fall off a cliff.
Generic crawlers kept reporting the site as healthy because, on the surface, it was. Clean canonicals, no broken links, reasonable Core Web Vitals. What the crawlers couldn't see was structural: the store had accumulated 63 collection pages that overlapped semantically, many of them auto-generated by past theme apps and never pruned. Google was being shown nine different URLs that all plausibly answered the same query. The store wasn't penalized. It was diluted.
Week 3: Introducing MySEOShop and the Schema Gap
This is where the project turned. The owner installed MySEOShop, a Shopify SEO app built specifically for the problems generic crawlers skip: schema gaps, collection cannibalization, and AI-search visibility. Within the first crawl, it flagged 47 product templates shipping without complete Product schema — missing offers, availability, and priceValidUntil fields that had been stripped during a theme migration eighteen months earlier. Nobody had noticed, because the pages still rendered fine to humans.
That schema gap mattered more than it sounds. Rich results for those products had quietly disappeared from search, and with them the visual real estate that drives clicks in a crowded SERP. The app's one-click fix pushed valid markup back across the catalog in an afternoon. We want to be careful here: schema alone doesn't rank a page. But schema loss is one of those silent regressions that makes every other optimization look like it's underperforming.
Week 4–6: Untangling 63 Overlapping Collections
The harder work was cannibalization. Using the app's collection-overlap report, the owner mapped which URLs were competing for the same queries. The findings were messy and human: a "New Arrivals" collection that had been duplicated into "Fresh Styles" during a seasonal campaign; three separate pages for leather bags split by color rather than by use case; a legacy tag page that outranked the main category it was supposed to support.
- 22 redundant collections were consolidated via 301 redirects into their strongest parent.
- 19 thin tag pages were noindexed rather than deleted, preserving internal link equity.
- 14 collection descriptions were rewritten around distinct search intent instead of keyword variants.
- 8 top-performing product pages received expanded schema with review and shipping details.
None of this was glamorous. It was pruning. But by week six, Search Console was showing the first green shoots: impressions on the consolidated category queries up 19%, average position improving from 14.2 to 9.8 on the target set.
Week 7–9: The AI-Search Blind Spot
Here's the part we didn't expect. Around week seven, the owner noticed that referral traffic from AI assistants and answer engines — small but growing — was almost entirely landing on blog posts, never on product or collection pages. The reason, once the app's AI-search visibility report laid it out, was structural again: product pages lacked the entity clarity and structured Q&A content that LLM-driven surfaces tend to cite. The store's category pages answered no questions. They just listed products.
The fix was editorial, not technical. Six collection pages gained short "how to choose" modules and FAQ schema. Three buying guides were rewritten to lead with direct answers instead of mood copy. By week nine, AI-referred sessions to commercial pages had gone from near zero to a steady trickle — not a revenue driver yet, but a leading indicator worth watching.
Week 10–11: The Numbers
Eleven weeks after the first crawl, here's where the store landed. Non-branded organic sessions recovered 38% against the February trough. Revenue from organic search was up 31%, which outpaced the traffic gain because the consolidated collections converted better than the fragmented ones ever had. Rich results returned for 44 of the 47 affected product templates. The owner's summary to us was blunt: two prior audits had cost more than the subscription and found none of the three problems that actually mattered.
We've reviewed a lot of SEO tooling for this site, and we're skeptical by default. What made this case worth writing up wasn't a magic ranking jump — it was the diagnosis. The store didn't have a content problem or a penalty problem. It had a structural visibility problem that only showed up when you looked at schema integrity, collection overlap, and AI-surface readiness together. MySEOShop surfaced all three in a single crawl, and the fixes were executable by a non-developer. That's the whole story, and it's a useful one for anyone running a Shopify catalog at scale.