Tripling Organic Traffic for a Jordanian Home Goods Retailer

Zaytouna Home sold furniture and homeware across Jordan with a catalogue of roughly 4,000 products. Their English pages ranked reasonably. Their Arabic pages — which served the majority of their actual customers — ranked for almost nothing.

The Arabic site had been produced by machine translation two years earlier and never revisited. There were no hreflang tags, so Google was indexing one language version and treating the other as duplicate content. Product descriptions read as translations because they were. And the Arabic pages loaded noticeably slower than the English ones because the RTL stylesheet was loading on top of the LTR one rather than instead of it.

Commercially, the effect was that paid advertising was carrying almost all of their Arabic-language revenue, at a cost per acquisition that was rising every quarter.

This Arabic SEO case study sets out the brief, the approach and the measured result.

This Arabic SEO case study covers eight months of work for a home goods retailer, from technical remediation through to the content rebuild that tripled their organic traffic.

We ran the programme in three phases over eight months, sequenced so that technical fixes landed before content work — there is no point writing pages Google cannot index correctly.

Phase one: technical foundation (weeks 1–6)

We implemented reciprocal hreflang tags across all 4,000 product pages plus an x-default, and corrected the canonical tags, which had been pointing every Arabic page at its English equivalent — effectively asking Google to drop the Arabic site from the index.

The stylesheet loading was rebuilt so RTL pages load only the RTL bundle. That alone took 1.4 seconds off Arabic page load. We also replaced fourteen physical CSS properties with logical equivalents, which fixed layout breakage on mobile that had been suppressing Arabic mobile rankings.

Phase two: keyword and content rebuild (weeks 5–20)

We rebuilt the Arabic keyword set from scratch rather than translating the English one. This surfaced the single largest finding of the project: their customers searched using colloquial Jordanian terms and Arabic-script transliterations of English product names, while every page on the site used formal Modern Standard Arabic.

The term for a three-seater sofa that customers actually typed had roughly eleven times the search volume of the formal term the site was optimised for.

We rewrote the 180 highest-value category and product pages in natural Arabic, and built twenty-two buying guides targeting informational queries that had no good Arabic-language result in Jordan at all.

Phase three: authority and local signals (weeks 12–32)

We standardised the business name, address and phone across Jordanian directories in both scripts, rebuilt the Google Business Profile with Arabic as the primary language, and earned links from local interior design publications and two university design programmes.