Sources, method and limits, stated in full

How We Read the Dubai Land Department Register

Every statistic on this site comes from the Dubai Land Department's own open data — not from developer marketing, not from portal listings, and not from estimates. This page exists so that anyone, including a journalist or an analyst who wants to quote us, can see precisely how a number was produced and reproduce it themselves. It also sets out plainly what this data cannot tell you, because a source that only lists its strengths is not being straight with you.

Primary source
DLD open data
gateway.dubailand.gov.ae
Government register
Update frequency
Daily
Snapshots retained and dated
Every figure carries its date
2026 projects tracked
361
Full registration year
Recomputed, not copied
Transactions analysed
70,940
Recorded off-plan purchases, 2026
Across 1,378 developments

Where the data comes from

We use two Dubai Land Department open-data endpoints. The first is the project register, which records every off-plan project filed with the DLD: developer, area, declared value, registered unit count, escrow account status, construction percentage and the contractual completion date. The second is the transaction register, which records completed property sales: price paid, area, unit size, room count, property type and whether the sale was off-plan. Both are public, both are published by the Dubai Land Department, and neither requires a commercial licence to read. We query them directly rather than through an intermediary, take a dated snapshot each day, and retain those snapshots so that any figure we have published can be traced back to the exact dataset it came from. Nothing on this site is scraped from a portal, supplied by a developer, or estimated where the register is silent.

Two DLD open-data endpoints, queried directly, snapshotted daily and retained.

How the numbers are calculated

Project counts are straightforward: we count rows in the register for the relevant year, filtered by developer or area where stated. Unit counts and declared values are summed from the register's own fields. Escrow status is read directly from the register's escrow flag — we do not infer it. Transaction figures need slightly more care. We aggregate every recorded off-plan sale by project name, then report the median price rather than the mean, because a handful of penthouse sales will drag a mean upward and misrepresent what a typical buyer paid. Price per square foot is calculated per transaction from the recorded value and the recorded area, converted from square metres, and then the median of those is taken — not the median price divided by the median size, which would give a different and less honest answer. Where a development is registered in phases (Azizi Venice runs to sixteen), we aggregate the phases and say so on the page. Sibling projects are matched on significant word tokens, never on substrings, because substring matching produced a genuine false positive during development: "Avelia at The Valley" contains "lia" and would otherwise have absorbed an unrelated project called LIA.

Medians, not means. Per-transaction price-per-sqft, then the median of those. Phases aggregated and disclosed.

What this data cannot tell you

Four real limits, and they constrain what any honest analysis of this source can claim. First, the open data covers the current registration year only. Requests for 2025 and earlier return an empty result, so we cannot show multi-year trends, and any site claiming a DLD-sourced historical series is not using this source. Second, every off-plan transaction carries a single procedure type, so a developer's first sale cannot be distinguished from a resale or an assignment. When we report transaction counts we are counting all off-plan sales, not new sales, and we say so. Third, unit counts are missing on some registered projects — villa and townhouse clusters frequently register with a zero unit count — so a total that sums registered units will understate true supply in villa-heavy districts. Fourth, the register reflects filings, not reality on the ground: a completion date is the date filed, a construction percentage is the percentage reported. We treat both as claims made to a regulator, which is more reliable than marketing but is not the same as verified fact.

Current year only. No developer-sale-vs-resale split. Some unit counts absent. Filings, not site inspections.

The escrow figure, explained

The number we are asked about most is that 102 of the 359 projects registered in Dubai in 2026 — 28% — had no open escrow account at the time of checking. Here is exactly what that means. Under Dubai's off-plan framework, a developer must open a project escrow account before units in that project can lawfully be sold; buyer payments go into that account rather than to the developer directly. The register carries a flag for whether the account is open. We count the rows where that flag is false. That is the entire calculation. What the figure does not mean is that 28% of Dubai projects are fraudulent or in breach. Escrow accounts commonly open within weeks of a project being registered, and the interval between registration and escrow is ordinary administrative sequencing. The reason the number matters is that marketing does not wait for it: brochures, prices and floor plans routinely circulate before the account opens, and a buyer has no way of knowing from that material which side of the line a project sits on. The figure is a snapshot and it moves. We date every publication of it, and anyone can check a specific project themselves against the register.

28% is a snapshot of a moving figure, and it is not an accusation. It is a reason to check before paying.

How to verify anything we publish

Take any figure from this site and check it. The Dubai Land Department publishes the underlying data openly, and every number we report carries the date of the snapshot it came from. For a specific project, the fastest route is to ask the developer or broker for the project number and escrow account number in writing, then confirm both against the DLD register — a screenshot supplied by a broker is not verification. For aggregate figures, the endpoints named above return the same rows to anyone who queries them. If you find a discrepancy between something we have published and what the register currently says, the most likely explanation is that the register has moved since our snapshot; the second most likely is that we have made an error. We would rather hear about the second. Corrections are made on the page itself with the change noted, not quietly.

Ask for the escrow account number in writing and check it yourself. A broker's screenshot is not verification.

Independence

This site earns no commission on any property transaction. It does not list units for sale, does not represent developers, and is not paid for coverage or placement. That matters to the method, because the findings that make this data worth publishing — escrow accounts that are not open, registered completion dates years later than the marketed ones, advertised entry prices well below what buyers actually pay — are precisely the findings a site earning commission has no incentive to surface. Where we get something wrong, we correct it. Where the data does not support a conclusion, we say the data does not support it rather than filling the gap with an opinion.

No commission, no listings, no paid placement. That is what makes the uncomfortable findings publishable.

Our verdict

Every figure on this site comes from Dubai Land Department open data, queried directly, snapshotted daily and dated on publication. Medians are used rather than means, phased developments are aggregated and disclosed, and the limits — current year only, no resale split, some missing unit counts, filings rather than inspections — are stated wherever they affect a conclusion. Any number here can be reproduced from the same public source.

Frequently Asked Questions

Where does Certified Poor get its Dubai property data?

From two Dubai Land Department open-data endpoints: the project register (developer, units, escrow status, construction progress, completion dates) and the transaction register (recorded sale prices, areas, unit sizes). Both are public government data. We query them directly and retain dated snapshots so any published figure can be traced to its source.

How often is the data updated?

Daily. A snapshot is taken and retained each day, and every figure we publish carries the date of the snapshot it came from. The register itself updates continuously as projects register, escrow accounts open and construction progresses, so a figure quoted without a date is not meaningful.

What does the 28% escrow figure actually mean?

That 102 of the 359 projects registered with the DLD in 2026 had no open escrow account at the time we checked. Escrow is the legal gate before off-plan units can be sold and the mechanism that protects buyer deposits. It is not an allegation of wrongdoing — accounts commonly open within weeks of registration. It matters because marketing frequently begins before the account opens.

Can I quote or cite these figures?

Yes. Please cite the snapshot date alongside the figure, because the register moves. The underlying data is public, so any number here can be independently reproduced from the Dubai Land Department's own endpoints. If you find a discrepancy we would rather you told us.

Why do you use medians rather than averages?

Because a small number of penthouse or villa sales pulls a mean upward and misrepresents what a typical buyer paid. On a development with 6,000 studio sales and 20 penthouse sales, the mean describes nobody. Price per square foot is also calculated per transaction and then median-ed, rather than dividing a median price by a median size, which would produce a different and less accurate figure.

What are the limits of this data?

Four. The open data covers the current registration year only, so no multi-year history. Off-plan transactions carry one procedure type, so developer sales cannot be separated from resales. Some registered projects — mostly villa and townhouse clusters — carry a zero unit count, which understates supply in those districts. And the register records what was filed with the regulator, not what an inspector observed on site.

Is Certified Poor paid by developers?

No. The site earns no commission, lists no units for sale, represents no developer and takes no payment for coverage or placement. That independence is what makes it possible to publish findings — missing escrow accounts, completion dates later than advertised, prices above the marketed entry point — that a commission-earning site has no reason to surface.