Methodology

How the Rendr Score works

Every off-plan project with a price per square foot gets one number from 0 to 100 that says how well it stacks up, and every number shows its working. 8 factors, each with a weight, each scored by a rule written here; a handed-over building gets the Rendr Resale Score, from 5 factors (below), one of them its growth since launch. Some are computed from public data; some are our own judgement, and where they are, the reasoning and sources sit next to the score on the project page. Our opinion, not investment advice.

The factors and their weights

Weights add up to 100. The total is the weighted average of the factors that could be assessed, on the Rendr scale (below). These weights apply since 1 Oct 2026.

1. Price against what buyers actually paid

   Automatic. The project's price per sqft against the emirate register's sales in the area (in Dubai, also the land district the register files the area's sales under, named in the working), apartments or villas, the developer's sales (Abu Dhabi) or off-plan sales (Dubai) of the last six months, widening to twelve months and then to every sale where there are fewer than thirty, freehold sales only, and no one project allowed more than the next project's count or a fifth of the figure. Every project is measured against the median of those sales; a project starting at AED 10M or more against the higher of the area's top quarter and the median of what buyers paid across the emirate for homes of AED 10M and more (the most expensive tenth of every sale when there are fewer than thirty of those). The project's figure is, in order: the median of its own sales in the register (developer sales in Abu Dhabi, off-plan sales in Dubai); the price per sqft the developer states; our estimate from the price and size of its typical home.

   Rule: 20% or more under the benchmark scores 100. From 20% under to the benchmark: 100 down to 70. Up to 15% over: 70 down to 50. Up to 40% over: 50 down to 20. Beyond that: 20 down to 0. When the figure is our own estimate it is labelled as such, its gap counts half (an estimate is typically 8% out), and it cannot score above 70: a guess can say a project is not obviously overpriced, never that it is a bargain. A project whose own registered sales are more than half of the sales behind its area's figure is measured against the median, whatever its tier: it is most of that figure itself. An estimate under half or over three times the benchmark is taken as a wrong figure in our sources, not a price: it is not used, and the project has no price factor until the figure is put right or a person has checked that it is real.

   25weight

2. Developer track record

   Four parts, each counted only where it has data: our assessment of each developer, with the reasoning and sources shown; the delivery-record rule over its typed facts (projects delivered in the last ten years, the typical delay, cancellations, any problem open today); how its finished buildings resell against their areas in the emirate's register; and researched facts with sources: who stands behind it, financial distress, and documented quality and after-sales cases. Under it, what the register counts under the developer's names.

   Rule: Our rating weighs 35, the delivery-record rule 30, resale performance 20 and the researched facts 15, averaged over the parts that have data. Our rating: 90 and above, many projects delivered on time; around 70, delivers with delays of a few months; around 50, little history or delays over a year; below 30, unresolved problems. The rule: many delivered on time scores up to 92; three months' grace, then two points a month late to a year and one and a half after, at most 35; eight a cancellation; a problem open today holds it at 25. Resale performance: each finished building's own resale median (ten sales or more) against its area's finished-home sales (thirty or more), over two buildings or more, 60 at par and 1.5 points a percent, pulled toward 60 when few buildings stand behind it. Where no typical delay is on file, the register stands in for the rule: of the developer's projects with a registered completion date, the share past it and not finished, from 92 with none down to 35. Researched facts: government-backed 85, listed 80, private with ten or more delivered 70, fewer 55; 10 off a documented quality case (at most 30), 5 off an after-sales dispute (at most 15); financial distress in the last five years holds it at 25. The factor applies to every project of that developer.

   20weight

3. Payment plan

   Automatic, from the plan's percentages.

   Rule: 100 minus the down payment, minus 0.6 times the share due during construction, plus 0.4 times any share due after handover (counted up to 30% of the price), kept between 0 and 100. A 10/50/40 plan scores 60; a 5/35/60 plan scores 74; paying everything on handover scores 100.

   10weight

4. Location and infrastructure

   Our assessment of each area, with the reasoning and sources shown (a district inherits its island's rating unless rated on its own; a project can be rated on its own), and what the project's own pin is near, measured on OpenStreetMap: a metro or tram stop (Dubai), a school, a hospital or clinic, a supermarket or mall, a beach.

   Rule: Our rating counts 60%: schools, healthcare, shops, beach, roads, and how much of the area's infrastructure is finished, on a 0 to 100 scale. The distances count 40%: each place scores 100 when near and 0 when far (a stop within 800 m to 3 km, a school 1 to 5 km, healthcare 2 to 10 km, shops 500 m to 3 km, a beach 500 m to 5 km), weighted 20, 20, 20, 25 and 15; Abu Dhabi has no metro, so that place is left out there rather than counted as far. What is not built or not mapped yet does not count. A villa or townhouse project is measured against its emirate's villa and townhouse projects: its place among them takes the score an apartment project scores at the same place, since a house community sits away from the metro by design.

   15weight

5. How big the resale market is

   Automatic. Resales of finished homes of the same kind registered in the area, counted over the last twelve months (or since 1 January where the register reaches no further) and turned into a yearly figure.

   Rule: On a log scale: 10 apartment resales a year score 25, 100 about 58, 1,000 about 90, and it tops out near 2,000 (villas, fewer and larger homes, from 4 a year). Where no finished home of the kind has resold yet and four in five of the area's sales are still off-plan, there is nothing to resell: the factor is not assessed, and the supply factor reads the new community instead; any resale at all is measured, so a thin resale market scores low. An area with no sales of this kind at all is not assessed either. It measures the size of the resale market, not how fast one home sells. A villa or townhouse project is measured against its emirate's villa and townhouse projects, as for location; no resales recorded stays 0.

   10weight

6. Construction on schedule

   Automatic from the register's % built, the handover date first promised and the launch date, each reading where its data exists; otherwise our assessment.

   Rule: The lowest of three readings. On the curve: nothing is expected in the first nine months after launch (design, permits and enabling works; a quarter of the period when that is shorter), then progress on an S-curve to handover, measured on the day % built was read; up to ten points behind: 100, each further point costs three, so 43 behind is 0, and while under 10% is due it is too early to tell. Late: past the handover first promised and not finished, each month costs five points (70 at six months, 40 at a year), counted to the day % built was read, or to today when no % is on file. At pace: what is left, built at four points a month from that reading, finishing more than three months after that promise costs five points a month beyond it, the same test as the page's "likely later". A handover the register moved later is measured against the first one promised. When assessed by us: 100 on schedule, 70 up to six months late, 40 up to a year, 0 beyond.

   15weight

7. How much is still being built nearby

   Automatic. The share of the area's registered home sales over the last twelve months that were off-plan, apartments and villas together.

   Rule: 30% or less off-plan scores 100; the score falls evenly to 10 at 95% or more. A market that is mostly off-plan has many homes still to be handed over against the finished ones that trade today, and a wave of handovers weighs on prices and rents.

   5weight

8. The basics on file

   The emirate's own register: the project's registration and its escrow account. A registered off-plan project sells through escrow by law.

   Rule: Shown beside the score, not counted in it (weight 0 since 1 Oct 2026): a registered off-plan project sells through escrow by law, so a number on file mostly says we matched it, and it could only add points. A registration or escrow number on file: on the register. Checked by hand and none found: said so.

   0weight

One rule over the average

The Rendr scale

The total is the weighted average on the Rendr scale, a fixed curve: about 7 points more for a typical project, none for the weakest, the top pressed towards 100. Averaging eight factors pulls every project towards the middle; the scale keeps the middle of the market from reading as a bare pass. It never changes the order, and a total moves only when an input does.

Straight lines between the rows; below 40 (off-plan) or 45 (resale) the average shows as it is.

The Rendr Resale Score, for finished buildings

A handed-over building has nothing left to deliver, so it gets a resale score instead of the off-plan one: its own registered price against the area's, how much its registered price has grown since launch against buildings launched the same year, the location, how easily finished homes resell, and the developer. No construction, payment plan, supply or paperwork factor, and no developer cap. It is a different score with its own name, so the two are never ranked or compared against each other. These weights apply since 3 Oct 2026.

1. Price against what buyers actually paid

   Automatic. The building's own sales of finished homes, at least ten (resales in Abu Dhabi, finished sales in Dubai; every sale where too few), against the area's sales of finished homes of the same kind (every sale in the area where those are under thirty). A building just handed over with no resales yet keeps its own off-plan sales against the area's, like with like, until ten resales register. A launch price on a finished building is years old, so a building without its own registered sales is not priced.

   Rule: The same curve as the off-plan score: 20% or more under the benchmark scores 100, at the benchmark 70, 15% over 50, 40% over 20.

   30weight

2. Growth since launch

   Automatic, finished buildings only. Dubai: the Dubai Land Department's open data (data.dubai), every registered sale since 2009. Abu Dhabi: ADREC's register of sales (DARI), since 2019. The building's own registered sales in the first twelve months of its off-plan sales (the developer's, in Abu Dhabi), against its resales of finished homes in the last twelve months, apartments or villas.

   Rule: The change from the launch year's median price per sqft to the resales' median, as a yearly rate, against the median yearly rate of the emirate's measured buildings of the same kind launched the same year (the same year with both kinds, or the whole emirate, where fewer than eight), so a building is not marked up or down for when it launched. Level with them scores 60; each percentage point a year above or below moves it four points, between 0 and 100. Needs ten sales in the launch year, five resales in the last twelve months and two years since launch; a launch price under half its area's off-plan median that year is a recording slip and not counted. Launches before the register's data start (2009 in Dubai, 2019 in Abu Dhabi) are not assessed.

   20weight

3. Developer track record

   As in the Rendr Score, above.

   15weight

4. Location and infrastructure

   As in the Rendr Score, above.

   20weight

5. How big the resale market is

   As in the Rendr Score, above.

   15weight

What the number means

• Strong80 to 100
• Good65 to 79
• Fair50 to 64
• Weakbelow 50

Missing data

What the score claims

Confidence

• HighThe price factor rests on the project's own registered sales, at least 30 of them, against an area median of at least 100 registered sales.
• MediumThe price factor rests on fewer registered sales of the project (the floor is 10), or on the price per sqft the developer states, or on an area median of fewer than 100 sales (the floor is 30); or the price evidence is solid but the developer's record is not on either register under the names we hold, so the developer rating rests on the developer's own pages and the press.
• LowThe price factor rests on our own estimate from the price and size of the project's typical home. Its gap counts half and the factor is capped at 70 in that case, and the score should be read as provisional until the register shows the project's sales.

Corrections

Changes to the weights

Every change is kept. A project page explains a score change by the factor that moved, and old scores are read against the weights that applied at the time.

• 1 Oct 2026current

  Price 25 · Developer record 20 · Payment plan 10 · Location 15 · Resale 10 · Construction 15 · Supply 5

  Score v13 (2026-10-01, the audit): basics 5 to 0 (it could only add points; shown beside the score), construction 10 to 15 (the curve, months past the first promised handover, the build pace). docs/score-design.md "v13"

• 26 Sep 2026

  Price 25 · Developer record 20 · Payment plan 10 · Location 15 · Resale 10 · Construction 10 · Supply 5 · Paperwork 5

  Score v2 (2026-09-25): supply added at 5; payment plan 15 to 10 (plans barely differ and moved no ranking). docs/score-design.md "Score v2"

• 8 Sep 2026

  Price 25 · Developer record 20 · Payment plan 15 · Location 15 · Resale 10 · Construction 10 · Paperwork 5

  Initial weights, docs/score-design.md, agreed 2026-09-08

Questions about a score? The breakdown on each project page shows every input. The registered-sale figures behind two of the factors are explained on the home page.