Skip to content
Get Ranked On TopIllustrative case studySignal plan

A fictional business. The figures are a modelled scenario, not a client result.

Rayo Lavandería · 11-branch laundromat chain · Medellín

How a Medellín laundromat chain washed out its duplicate pages and got all 11 branches indexed

All 11 branches indexed, +64% organic pickup bookings

8 months · Latin America · Signal plan

  • Duplicate consolidation
  • JavaScript rendering
  • HTTPS migration
  • Sitemaps & schema

Organic pickups · month 8

01The dirty load

Eleven branches, one page wearing eleven name tags.

Branch pages were copies with the names swapped, the pickup app hid hours and prices from Google, and every URL answered on both http and https.

Rayo grew from one laundromat in Laureles to eleven branches across the Aburrá Valley, and the website grew the cheap way: one branch template, cloned, with the barrio name swapped in. A marketing push then added 20 “laundry near me” barrio pages and 6 municipality pages built the same way. Meanwhile the new pickup app, which also fed opening hours and prices into every branch page, sat behind a robots.txt block, and the 2019 move to HTTPS had never been finished. Google saw a pile of near-identical, half-empty pages, two copies of each, and indexed three of them.

1 · Swapped names

37 pages shared 94% of their text. Only the branch or barrio name changed.

Two branch pages, as published

/sedes/belen

Rayo Lavandería Belén: tu lavandería autoservicio de confianza. Lavamos, secamos y doblamos tu ropa con recogida a domicilio en Belén y alrededores.

/sedes/robledo

Rayo Lavandería Robledo: tu lavandería autoservicio de confianza. Lavamos, secamos y doblamos tu ropa con recogida a domicilio en Robledo y alrededores.

2 · Blocked app

robots.txt blocked /app/, which also served the script and API behind every branch page’s hours and prices.

What Google rendered

User-agent: *
Disallow: /app/
Disallow: /*?sede=
Cargando sede…

3 · Two copies of everything

Every URL answered 200 on http and on https, with no redirect and no HSTS. 2,340 http URLs had made it into the index.

One page, two addresses

  • http://rayolavanderia.example/sedes/belen200
  • https://rayolavanderia.example/sedes/belen200

02Care label

What the audit found sewn into the seams.

Duplicate content

94%

A similarity crawl of all 37 location pages: 94% shared text on average, identical titles apart from the name, and Search Console flagging 29 of them as duplicates where Google had chosen another canonical.

Finding 1 of 4

Rendering

8 / 11

In the URL Inspection render, 8 of 11 branch pages showed only a loading spinner: their hours, prices and services came from /app/api/sedes, which robots.txt disallowed.

Finding 2 of 4

Protocol

2,340

2,340 http URLs indexed alongside their https twins. 41% of internal links were hard-coded to http, and the sitemap listed both versions.

Finding 3 of 4

Crawl waste

58%

Eight weeks of server logs: 58% of Googlebot requests went to http duplicates and ?sede= parameter URLs. A branch page was revisited once every 19 days on average.

Finding 4 of 4

robots.txt · Before · month 0
User-agent: *
Disallow: /app/
Disallow: /*?sede=
robots.txt · After · month 8
User-agent: *
Disallow: /*?utm_
Sitemap: https://rayolavanderia.example/sitemap.xml

03The programme

Four settings on one dial.

Built on our Signal plan: the technical engagement for sites where crawling, rendering and indexing decide what customers can find.

  1. Duplicate consolidation

    The 20 barrio pages were merged into the branch that actually serves each barrio, with one-hop 301s. The 11 branch pages and 6 municipality hubs were rewritten from real data: machines and load sizes, pickup zones, parking, opening hours, photos of each shop. Every page that stayed got a self-referencing canonical.

    Fixes the Spin cycle

  2. JavaScript rendering

    robots.txt now lets Googlebot fetch the app’s scripts and API, while the booking flow itself stays out of the index with noindex. Hours, prices and services are rendered on the server into the branch page HTML, and the booking widget loads after it.

    Fixes the Rinse cycle

  3. HTTPS migration

    Every http URL 301s to its https twin in one hop, HSTS is on, and 1,960 hard-coded internal links were rewritten. Canonicals, the sitemap and all 11 Google Business Profile website links now point at https.

    Fixes the Wash cycle

  4. Sitemaps & schema

    One XML sitemap of 18 clean URLs, with lastmod taken from the branch database. LocalBusiness schema on every branch page with opening hours, coordinates and the pickup service area, validated in the Rich Results Test.

    Fixes the Wash cycle

04Run the load

Wash, rinse, spin: watch the dirty URLs come out in the filter.

Twelve URLs from the site go into the drum. Wash is the crawl, rinse is the render, spin is the index. Run the load as the site was before the work, then as it is after, and see what comes out clean.

Arrow keys turn the dial.
123456789101112

Twelve URLs from the site, as the audit found them.

Drum loaded with 12 URLs. Turn the dial or run the full programme.

In the drum 12

  • 1/sedes/el-poblado
  • 2/sedes/laureles
  • 3/sedes/envigado
  • 4/sedes/belen
  • 5/sedes/robledo
  • 6/sedes/sabaneta
  • 7/sedes/bello
  • 8/lavanderia-medellin
  • 9http/sedes/belen
  • 10/lavanderia-cerca-de-mi/estadio
  • 11http/lavanderia-cerca-de-mi/florida-nueva
  • 12/app/reservar?sede=bello

Caught in the filter 0

Nothing yet.

Out clean · indexed 0

Finish the spin to see what comes out clean.

Indexed from this load: 0 of 12Before the work: 3 · after: 8

05Month by month

8 months, one full programme.

  1. 1 · Month 1

    Sort the load

    Similarity crawl of 37 location pages, eight weeks of logs, render tests of every branch page and a URL inventory across http and https.

  2. 2 · Month 2

    One address

    http to https in one hop, HSTS, 1,960 internal links rewritten and the sitemap cut back to https URLs only.

  3. 3 · Months 3–4

    Let Google in

    robots.txt opened for the app’s scripts and API; hours, prices and services rendered on the server; the booking flow set to noindex.

  4. 4 · Months 5–6

    One page per branch

    20 barrio pages merged into their branches with 301s; 11 branch pages and 6 municipality hubs rewritten from branch data.

  5. 5 · Months 7–8

    Fold and stack

    LocalBusiness schema, a clean 18-URL sitemap, Business Profile links moved to https and a weekly index check per branch.

06Results

Out of the spin: eleven clean branch pages, all indexed.

Organic pickup bookings / month

+64%

410 → 672 pickups a month from organic search

Branch pages indexed

3 → 11 of 11

Three were indexed before the work; the last joined in month 7.

http URLs in Google's index

2,340 → 0

Every http URL now 301s to https in one hop.

Near-duplicate pages

37 → 0

Highest similarity between two pages: 94% → 31%

Modelled scenario for a fictional business — not a client result. Rankings and traffic are never guaranteed; see our Terms.

Organic pickup bookings per month
0200400600M0410M1M2M3M4M5M6M7M8672
http URLs in Google’s index
2,3400M0M2M4M6M8

Googlebot now revisits a branch page every 3 days; it used to be every 19.

Branch board: the month each page came out indexed

Get a clean cycle for your site

Branch pages that look alike, an app Google cannot see, a half-finished move to HTTPS? Start with the Signal plan, or tell us what is in your drum.

← All case studies