Skip to content
Get Ranked On TopIllustrative case studyCommerce SEO plan

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

WASM CAMEL DAIRY · Camel-milk dairy & online shop · Al-Kharj

How a camel dairy near Riyadh got Arabic searches landing on its Arabic pages

+164%

organic revenue from the Arabic store

Al-Kharj, Saudi Arabia · 12 months · Retail & e-commerce

Every price on the board is written twice: ٣٥ and 35 are the same number in two scripts. The shop’s 120 products were written twice too, in Arabic and English, and nothing told Google the two pages were the same product. Modelled data.

THE CHALLENGE

One counter, two languages, and Google served the wrong one.

Arabic and English product pages duplicated each other with no hreflang, and delivery-zone filters trapped the crawler, so English pages won Arabic searches.

Wasm sells fresh camel milk, laban, cheese and chocolate from its farm in Al-Kharj, with same-day delivery across Riyadh. Most customers search in Arabic. Yet for “حليب نوق” and almost every other Arabic product query, Google showed the English page: same photo, same price, wrong language. Shoppers bounced back to the results and bought from whoever answered them in Arabic.

Two faults sat underneath. The Arabic and English catalogues were exact twins with no hreflang between them, so Google treated them as duplicates and kept the English one. And the delivery-zone filter turned every product list into thousands of crawlable URLs, so Googlebot spent its visits on filter combinations instead of the Arabic pages.

AUDIT NOTES · MONTH 0

  1. 0

    No hreflang anywhere

    240 product URLs, 120 in each language, and not one alternate annotation linking a page to its twin.

  2. 46,200

    Filter URLs crawled each month

    Fourteen delivery zones times sort orders times pack sizes: every combination was a link Googlebot could follow.

  3. 71%

    Arabic searches landing in English

    Seven in ten visitors who searched in Arabic arrived on an English page and saw prices in a script they did not shop in.

  4. 79

    Arabic pages folded into English

    Search Console reported them as duplicates where Google had chosen a different canonical: the English twin.

  5. 3,400

    URLs in one mixed sitemap

    Both languages and the filter combinations were listed together, with no sign of which page belonged to which language.

  6. 64

    Arabic pages with English titles

    The Arabic template kept the English title tag, which made the two versions look even more alike to a crawler.

THE PAIRING BOARD

Match every Arabic page to its English twin.

This is the fix in miniature. Each Arabic product page needs to name its English twin with hreflang, and the twin has to answer back. Pair the cards, then add the two final tags, and watch which page Google shows for an Arabic search.

ARABIC CATALOGUE/ar/

ENGLISH CATALOGUE/en/

Tip: the prices match. ٣٥ on the Arabic card is 35 on the English one.

Pairs: 0 of 6

6 duplicates left

Pick any card to begin, Arabic or English.

The <head> of /ar/p/camel-chocolate-dates

<!-- no canonical: Google may pick the English URL for both --><!-- no hreflang: Google sees two copies of one page --><!-- no x-default: no fallback for other languages and regions -->

ARABIC SEARCH PREVIEW

Searched for شوكولاتة حليب النوق

wasmdairy.example › en › p

Camel Milk Chocolate with Dates

SAR 35

Google shows the English page: wrong language for this shopper.

THE WORK

Four fixes, chalked up in order.

The engagement ran on our Commerce SEO plan: catalogue structure first, then crawl control, then the Arabic pages themselves.

  1. 240 PAGES PAIRED

    HREFLANG PAIRING

    Every page names its twin

    Hreflang pairs for all 120 products, ar-SA and en-SA, generated from the product ID so a new SKU is paired the day it goes live. x-default points to the English catalogue for everyone else.

  2. 79 PAGES RECOVERED

    SELF-REFERENCING CANONICALS

    Each page is its own original

    Self-referencing canonicals on both languages, so Google stops folding the Arabic page into the English one. Two sitemaps, one per language, with the alternates listed.

  3. -96% WASTED CRAWL

    FACETED NAVIGATION CONTROL

    Delivery zones out of the crawl

    The zone filter moved to a postcode check in the basket. Old filter URLs return the clean category with a canonical, and parameter links are no longer crawlable anchors.

  4. 64 TITLES REWRITTEN

    NATIVE ARABIC PRODUCT PAGES

    Arabic written for Arabic shoppers

    Native titles, descriptions and Product schema in Arabic, prices in the digits customers expect, and delivery answers for Riyadh neighbourhoods on each page.

TWELVE MONTHS ON THE BOARD

Rewritten one column at a time.

  1. MONTH 1SEPT

    Audit and crawl map

    Server logs, Search Console and a full crawl. Filter URLs and duplicate clusters mapped for every product.

    Arabic landing
    31%
    Indexed
    47/120
  2. MONTH 2OCT

    Zone filter closed

    Delivery zone moved to the basket; parameter links made non-crawlable. Googlebot’s filter visits halve.

    Arabic landing
    44%
    Indexed
    63/120
  3. MONTH 3NOV

    Hreflang pairs live

    All 120 products paired in both directions, generated from product IDs. x-default added.

    Arabic landing
    58%
    Indexed
    84/120
  4. MONTH 4DEC

    Self-canonicals

    Each page now names itself as canonical. Split sitemaps submitted, one per language.

    Arabic landing
    69%
    Indexed
    98/120
  5. MONTH 5JAN

    Arabic titles rewritten

    64 product titles and descriptions written natively in Arabic, not translated from the English.

    Arabic landing
    77%
    Indexed
    107/120
  6. MONTH 6FEB

    Ramadan gift season

    The gift box page in Arabic reaches the first page ahead of the season. Arabic revenue up sharply.

    Arabic landing
    83%
    Indexed
    112/120
  7. MONTH 7MAR

    Product schema in Arabic

    Prices, availability and delivery details marked up on both catalogues.

    Arabic landing
    87%
    Indexed
    115/120
  8. MONTH 8APR

    Neighbourhood answers

    Delivery times for 14 Riyadh areas added to product pages, in Arabic first.

    Arabic landing
    90%
    Indexed
    116/120
  9. MONTH 9MAY

    Category pages

    Milk, cheese and chocolate categories rebuilt in Arabic with buying guides.

    Arabic landing
    92%
    Indexed
    117/120
  10. MONTH 10JUN

    Holding steady

    Nine in ten Arabic searches now land on Arabic pages. Monitoring only.

    Arabic landing
    93%
    Indexed
    117/120
  11. MONTH 11JUL

    New SKUs, paired on launch

    Six new chocolates go live, each paired and canonicalised automatically.

    Arabic landing
    94%
    Indexed
    117/120
  12. MONTH 12AUG

    Year-end count

    117 of 120 Arabic pages indexed, and Arabic-store revenue up 164% on the same quarter last year.

    Arabic landing
    94%
    Indexed
    117/120

TOTTING UP

Arabic searches, answered in Arabic.

+164%

Arabic-store organic revenue (last quarter, year on year)
306090SeptDecMarJunthousand SAR / month
  • prior year
  • campaign year

29% → 94%

Arabic searches landing on Arabic pages
0%50%100%29%94%0612

-96%

Filter URLs crawled per month
46.2K1.9K0612

41 → 117

Arabic product pages indexed (of 120)
  • indexed before
  • recovered
  • not yet

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

TRACKED SEARCHES · POSITION AND LANDING PAGE

Search positions and the catalogue Google showed, before and after
SearchMonthly searchesMonth 0Month 12Landing page
حليب نوق9,900143 · Up 11 placesEnglish → Arabic
حليب ناقة الرياض2,400111 · Up 10 placesEnglish → Arabic
جبن حليب النوق1,300192 · Up 17 placesEnglish → Arabic
شوكولاتة حليب الإبل880232 · Up 21 placesEnglish → Arabic
حليب نوق مجفف720274 · Up 23 placesEnglish → Arabic
camel milk Riyadh59082 · Up 6 placesEnglish → English
camel milk chocolate1,000165 · Up 11 placesEnglish → English

TWO LANGUAGES, ONE CATALOGUE

Get a plan like this

A bilingual shop where each language ranks for its own customers, and Googlebot spends its time on pages that sell.

← Back to all case studies