We put 40 questions a Dubai off-plan buyer actually asks to ChatGPT, Claude, Gemini and Perplexity, and recorded which developers and firms each engine names. 354 names came back - only 35 firms appear on 2 or more engines, and a handful dominate the channel buyers increasingly start with.
When a Dubai buyer asks AI about off-plan, the same names come back - Emaar, DAMAC, Sobha, Nakheel, Binghatti. Of the 354 distinct names extracted from the answers, only 35 developers and firms are named on 2 or more engines once duplicate name variants are merged and districts, individual projects and other non-firm entities are removed. The rest rarely surface at all.
Visibility = share of the 160 AI answers naming the firm, with name variants merged before counting (Sobha / Sobha Realty, DAMAC / DAMAC Properties, Nakheel / Nakheel Properties, Danube / Danube Properties, Ellington / Ellington Properties). Only developers and firms named on 2 or more engines are ranked; districts, individual projects and non-firm entities are excluded.
| # | Firm | Questions (of 40) | Answers (of 160) | Visibility |
|---|---|---|---|---|
| 1 | Emaar | 24 | 64 | 40.0% |
| 2 | DAMAC | 21 | 34 | 21.3% |
| 3 | Sobha | 14 | 32 | 20.0% |
| 4 | Nakheel | 14 | 21 | 13.1% |
| 5 | Binghatti | 9 | 19 | 11.9% |
| 6 | Engel & Völkers | 12 | 16 | 10.0% |
| 7 | Meraas | 12 | 16 | 10.0% |
| 8 | Knight Frank | 8 | 13 | 8.1% |
| 9 | Betterhomes | 7 | 12 | 7.5% |
| 10 | Omniyat | 9 | 11 | 6.9% |
| 11 | Ellington | 7 | 11 | 6.9% |
| 12 | Select Group | 7 | 9 | 5.6% |
| 13 | fäm Properties | 5 | 9 | 5.6% |
| 14 | H&S Real Estate | 5 | 9 | 5.6% |
| 15 | Metropolitan Premium Properties | 5 | 9 | 5.6% |
| 16 | Azizi | 6 | 8 | 5.0% |
| 17 | Danube | 4 | 8 | 5.0% |
| 18 | Christie's International Real Estate | 3 | 7 | 4.4% |
| 19 | Driven Properties | 5 | 6 | 3.8% |
| 20 | Allsopp & Allsopp | 4 | 6 | 3.8% |
| 21 | Gulf Sotheby's | 2 | 6 | 3.8% |
| 22 | haus & haus | 3 | 5 | 3.1% |
| 23 | Luxhabitat | 3 | 5 | 3.1% |
| 24 | D&B Properties | 2 | 4 | 2.5% |
| 25 | Provident | 2 | 4 | 2.5% |
| 26 | Aeon & Trisl | 3 | 3 | 1.9% |
| 27 | Aldar | 3 | 3 | 1.9% |
| 28 | Dacha Real Estate | 3 | 3 | 1.9% |
| 29 | Nshama | 3 | 3 | 1.9% |
| 30 | Savills | 3 | 3 | 1.9% |
| 31 | White & Co | 3 | 3 | 1.9% |
| 32 | Arada | 2 | 3 | 1.9% |
| 33 | McCone | 2 | 3 | 1.9% |
| 34 | John Taylor | 2 | 2 | 1.3% |
| 35 | Range International | 2 | 2 | 1.3% |
| + 319 more extracted names sit below the 2-engine bar or outside the firm set. Get the full breakdown & where you rank → | ||||
Among the 35 developers and firms AI names on 2+ engines in this probe, the median visibility is 4.4%; the top firm is named in 40% of answers. Districts and master-communities (Dubai Marina, Downtown Dubai, JVC, Dubai Creek Harbour), individual projects and hotel-branded residence lines are excluded from the ranking.
These directories, regulators and platforms surfaced in AI answers. They are not competing for the buyer - they are the third-party sources AI re-reads. Getting listed and well-reviewed on them moves your own visibility.
| Source | Type | Engines |
|---|---|---|
| bayut.com | Directory | 1 |
| reddit.com | Forum | 1 |
| legalclarity.org | Publication | 1 |
| immigrationstartguide.com | Publication | 1 |
| gulfnews.com | Publication | 1 |
| + more sources cited across the cohort. Get the full source list → | ||
When ChatGPT, Claude, Gemini or Perplexity answer a buyer's question, the response is assembled from six technical inputs. No single marketing discipline covers them all - which is why most firms never make the 35 that AI names on 2+ engines.
We cover all six layers - the audit, the fixes, and the ongoing visibility - so AI starts recommending you. One programme, every engine.
28 Labs measures how brands get recommended across ChatGPT, Claude, Gemini and Perplexity. We've analysed 11,000+ AI answers covering 100+ brands across 15 markets - a uniquely deep view of what AI actually recommends, and why.
Emaar, DAMAC, Sobha, Nakheel and Binghatti are named most across ChatGPT, Claude, Gemini and Perplexity. In total 35 developers and firms are named on 2 or more engines; the rest of the 354 extracted names rarely surface.
160 answers surfaced 354 distinct names, but after merging duplicate name variants and removing districts and non-firm entities, only 35 firms are named on 2 or more engines.
The most-cited sources are bayut.com, reddit.com, legalclarity.org - the third-party pages AI re-reads when it builds an answer.
40 buyer questions x 4 engines (ChatGPT, Claude, Gemini, Perplexity) = 160 answers, single snapshot dated 2026-06-20, retrieval-enabled probe. Snapshot 1 - June 2026; a second snapshot is planned. Citations reflect the retrieval / web layer; some recommendations draw on training data with no visible citation. Not yet a benchmark.
Method. 40 buyer questions x 4 engines (ChatGPT, Claude, Gemini, Perplexity) = 160 answers, single snapshot dated 2026-06-20, retrieval-enabled probe. Each developer and firm is scored on whether the engines name it across the questions a Dubai buyer actually asks. "Visibility" = share of the 160 answers naming it, with name variants merged before counting. Citations reflect the retrieval / web layer; some recommendations draw on training data with no visible citation.
Status. Snapshot 1 - June 2026. A second snapshot is planned. Do not quote as a benchmark until it lands.
Press & data requests: [email protected]