We ran 1,200 commercial-intent prompts across five AI engines, split evenly between English and Arabic, across six categories in Jordan, Saudi Arabia and the UAE. The English results were roughly what we expected. The Arabic results were not.
The finding
In English, the median commercial prompt returned four named brands, and in 78% of cases at least three of those brands had recognisable, well-resourced content operations behind them. The competition was real and the gaps were small.
In Arabic, the median prompt returned two named brands. In 31% of Arabic prompts, the engine named no regional brand at all, defaulting to international players or declining to recommend. In another 19%, the brands named were ones whose Arabic content amounted to a machine-translated site.
Put plainly: in half of the highest-intent Arabic queries we tested, the field was effectively open.
Why the gap exists
Arabic is not underserved because of the language. It is underserved because of a strategic habit. Most regional businesses built their digital presence in English first, targeting an audience that included decision-makers, investors and international partners. Arabic arrived later as a toggle, usually machine-translated, usually with broken typography and no unique content.
A translated page is a weak retrieval candidate. It has no original claims, no local data, no named author with credentials in that language, and frequently no correct structured data. Models can parse it. They have very little reason to cite it.
Meanwhile, a substantial and growing share of buyers across the Gulf and Levant research in Arabic, including, in the categories we tested, the person who actually makes the decision rather than the person who signs the contract.
What to do about it
Research Arabic demand from scratch rather than translating your English keyword list. The questions people ask in Arabic are not the same questions with different words: the framing, the anxieties and the comparison points differ.
Write natively, in the right register. Jordanian Arabic, Gulf Arabic and Modern Standard are three different choices and the default of formal MSA for everything is often wrong for consumer categories.
Deploy structured data on both language versions with correct hreflang and a coherent entity graph. Give the model an unambiguous statement of who you are in both languages.
And publish something only you can publish: a price benchmark, a timeline study, a survey of your own customers. Original Arabic data is currently the cheapest citation you can buy anywhere in digital marketing.
The takeaway
This window will close. The categories we tested in Saudi Arabia are already measurably more contested than they were eighteen months ago. The brands that build genuine Arabic content operations in the next two years will hold positions that later entrants will have to pay a great deal to challenge.