Variants
1,540 duplicates
1,540 of the 2,140 indexable product URLs were colour or strap variants of another page, with near-identical copy and no canonical. Google switched the “main” variant from one week to the next.
Finding 1 of 4
A fictional business. The figures are a modelled scenario, not a client result.
Relojería Ancla · Watch retailer & repair shop · Valparaíso
+187% organic revenue in 12 months
12 months · Latin America · Commerce SEO plan
Caseback view · the catalogue as a movement
Fully wound · month 12 · Organic revenue (month 12 vs. month 1)
+187%
IThe challenge
Colour and strap variants split 600 watches into 2,140 competing pages, sold-out models went 404 and a flat category list lost brand searches to marketplaces.
Relojería Ancla sells and repairs watches from a workshop on Cerro Alegre. Its online catalogue had grown to 600 watches, but every dial colour and strap had become a product page of its own, so the shop was competing with itself. Sold-out watches returned a 404 until they were restocked, losing their links each time, and nine flat categories gave Google no page to rank for a brand or for “reloj automático”. Marketplaces, with clean category pages and price and stock shown in the results, took those searches.
One watch, four URLs
Before: four pages competing for one model
After: one canonical parent
IIDiagnosis
1,540 duplicates
1,540 of the 2,140 indexable product URLs were colour or strap variants of another page, with near-identical copy and no canonical. Google switched the “main” variant from one week to the next.
Finding 1 of 4
410 404s
410 product URLs returned a 404 in a year when stock ran out, 86 of them with links from blogs and collector forums. A restocked watch came back on a new URL and started from zero.
Finding 2 of 4
9 categories
Nine flat categories (men, women, offers, steel…) and no page per brand or movement. For “reloj automático” the best Ancla page was a filter URL at position 31.
Finding 3 of 4
0 marked up
Not one product page carried Product markup. Marketplaces showed price, stock and ratings right in the results; Ancla showed a title and a snippet.
Finding 4 of 4
IIIThe strategy
Built on our Commerce SEO plan: the ecommerce engagement for catalogues where templates, stock and structure decide what ranks.
One parent page per watch model, with colour and strap chosen on the page. Variant URLs stay for ads and old links but are canonicalised to the parent, and dropped from the sitemap: 1,540 duplicates folded into 600 models.
Sold-out watches stay live: a “notify me when it’s back” button, the expected date and three similar watches in stock. Discontinued models get a 301 to their successor or their category. Nothing returns a 404 any more.
A three-level tree built from what people search: movement first (automatic, quartz, chronograph), then brand and use. 64 categories, each with its own copy and breadcrumbs, and filters that stay crawlable only where there is demand.
Product, Offer and AggregateRating markup on all 600 watches, generated from the stock system so price and availability are never stale. Merchant listings in Search Console checked every week.
The category tree
IVWind the crown
Each gear is one catalogue fix, in the order we shipped them. Wind the crown to engage the next; the dial shows category-page sessions and organic revenue in the month that fix had landed.
Arrow keys move between gears.
Left subdial: category-page organic sessions a month, in thousands. Right subdial: organic revenue a month, in millions of Chilean pesos.
Mainspring let down · no fixes yet · month 1
Mainspring let down
2,140 product URLs for 600 watches, 404s on every sold-out model, nine flat categories and no structured data. Wind the crown to fit the first gear.
VMonth by month
IMonths 1–2
A crawl of all 2,140 product URLs, a year of 404 logs, search demand per brand and movement, and revenue per category.
IIMonths 3–4
Parent pages and the on-page selector built, canonicals and sitemap rules shipped; the duplicates leave the index over six weeks.
IIIMonths 5–6
Sold-out pages kept live with alerts, 230 discontinued models redirected, the 86 linked URLs recovered.
IVMonths 7–9
64 categories launched in three releases, with copy, breadcrumbs and internal links from the repair guides.
VMonths 10–12
Product markup on every watch and merchant listings fixed, just as the December gift season begins.
VIResults
Organic revenue (month 12 vs. month 1)
+187%
CLP 6.2M → CLP 17.8M a month
Category-page organic sessions a month
2,900 → 9,400
64 category pages, each with its own copy
"reloj automático" position
31 → 4
12 months: first page from month 7
Duplicate variant URLs indexed
1,540 → 0
600 parent pages, one per model
Modelled scenario for a fictional business — not a client result. Rankings and traffic are never guaranteed; see our Terms.
| Query, as typed | Page | Month 1 | Month 12 |
|---|---|---|---|
| reloj automático | Category | #31 | #4 |
| reloj automático hombre | Category | #38 | #5 |
| relojes halvard | Brand | #12 | #1 |
| reloj de buceo acero | Category | #27 | #3 |
| correa de cuero para reloj | Straps | #45 | #8 |
| reparación de relojes valparaíso | Repair service | #6 | #1 |
Queries are shown as Chilean shoppers type them, in Spanish.
Variants, dead stock and a flat catalogue are the most common problems in a shop this size. Start with the Commerce SEO plan, or tell us what you sell.