背景
Dubai's off-plan real estate market is one of the most active in the world, with hundreds of new projects launching each year. Investors face real information asymmetry. Developers control the narrative, brokers are incentivised to sell, and reliable comparable data is fragmented across government registries, property portals, and financial data providers.
Oliva was built to close that gap. We ingest data from independent authoritative sources, normalize it, and produce two numbers for every project. The first tells you how strong the project is on its fundamentals. The second tells you how well that project fits your specific goals.
Section A
Oliva 评分
Oliva 评分衡量什么
The Oliva Score is a 0 to 100 rating of how strong a project is on its objective fundamentals. It is the same number for every investor. It does not change based on who is viewing it. We build the score from six dimensions covering financial value, market dynamics, location quality, developer trust, risk, and liquidity. Macro context (interest rates, GDP, inflation, currency) is computed nightly and shown separately as Market Context on each project page - we keep it out of the composite because macro factors move every Dubai project identically in a given quarter, so they don't differentiate one from another.
Financial Value
22% weightWhether the price is attractive relative to the area, the income the property can generate, and the flexibility of the payment plan.
Market Dynamics
20% weightThe health and momentum of the local market, read through transaction activity, supply and demand balance, and pricing trends.
Location Quality
20% weightThe strength of the physical surroundings, including amenity density, transit links, proximity to schools and healthcare, and sub-area premium.
Developer Trust
15% weightThe reliability of the developer behind the project, measured by lifetime delivery history, years active, cycle resilience, and licensing standing.
Risk Assessment
15% weightDownside exposure in the area, including price volatility, oversupply risk, and concentration of a single developer or unit type.
Liquidity and Exit
8% weightHow quickly an investor can exit through resale velocity, rental demand, and the health of the broader capital market.
数据从哪里来
We only use authoritative sources. Every data point that feeds the Oliva Score can be traced back to one of the registries or providers listed below.
The official registry for every real estate transaction in Dubai. We use DLD data for transaction prices, volumes, and area-level trends.
The Real Estate Regulatory Agency within DLD. We use RERA data for project registration status, escrow compliance, and developer licensing.
Dubai’s official rental contract registry. We use Ejari data to validate rental demand and observed yields at the area level.
FRED (US Federal Reserve Economic Data)
The St. Louis Fed’s public economic data service. We use FRED for global macro signals including treasury yields, volatility indices, and commodity benchmarks that affect Gulf markets.
Licensed market data providers
Paid feeds for project-level detail, unit pricing, payment plans, and developer information, cross-referenced against DLD and RERA to catch inconsistencies.
The official registry for service charges and owner-association governance, reported per building.
Dubai Health Authority (DHA) Sheryan
The licensing authority for Dubai healthcare facilities, used to measure licensed clinic and hospital density by area.
The emirate’s official statistics agency, used for community population counts refreshed annually.
Roads and Transport Authority (RTA)
The authority behind Dubai Metro, used to read station ridership density near each project.
Ministry of Education (MoE) / KHDA
The regulators of private education in Dubai, used for operating private schools with student counts and quality ratings.
Data is updated continuously as new transactions and filings arrive. When a dimension is recomputed, the project's Score reflects the new state on the next view.
最新信号带来什么
The biggest underwriting improvement is real service charges. Until recently, the return model deducted a flat 15% across every project to cover management and service charges. With the RERA Mollak registry we now compute the actual service-charge burden per building, or per area when the building has not yet been matched, which produces materially different net-yield figures from one project to the next and from one area to another.
Area demand signals, including population, metro ridership, health-facility density, and school quality, add real context to the location dimension and replace the coarse proxies we relied on before.
Owner-association governance presence and the mortgage share of recent sales inform regulatory and liquidity risk with ground-truth data rather than estimates.
评分如何计算
The scoring pipeline has four steps. We describe each step qualitatively. We do not publish exact coefficients or cutoffs, for reasons we cover in the next subsection.
Step 1. Normalization
We normalize raw data points so different metrics live on the same scale. A price per sqft and a rental yield cannot be added together directly. Normalization puts them on a common footing so dimensions can be aggregated consistently.
Step 2. Confidence weighting
Newer data and more complete data matter more. Each input carries a confidence level that reflects freshness and coverage. A recent transaction near the project contributes more than a distant or stale one.
Step 3. Dimension aggregation
Within each dimension, the normalized and confidence-weighted inputs are combined into a single dimension score from 0 to 100.
Step 4. Overall Score
The seven dimension scores are combined into a single Oliva Score from 0 to 100. Financial Value carries the largest weight at 20%, followed by Market Dynamics and Location Quality at 18% each. Developer Trust and Risk Assessment sit at 14%. Macro Context and Liquidity round out the score at 8% each.
无付费排名
Developers cannot sponsor listings, cannot pay for higher rankings, and cannot influence their Oliva Score. The methodology is algorithmic. A project with weak fundamentals shows a weak score regardless of marketing spend. This is the single most important commitment we make to investors, and we will not compromise on it.
我们发布什么、不发布什么
We publish the framework, the six dimensions, and our data sources because you need to trust the score. We do not publish exact weights, sub-metric coefficients, or feature engineering details, for two reasons. First, to prevent developers from gaming the score by reverse-engineering the formula. Second, to protect ongoing research that improves the score over time. The framework and data sources are enough to verify the work is rigorous. The exact recipe is not.
Section B
Preference Match
Preference Match 是什么
The Preference Match is a 0 to 100% score that sits on top of the Oliva Score. The Score tells you how strong a project is objectively. The Match tells you how well that same project fits you, the specific investor. Two investors looking at the same project see the same Oliva Score but different Preference Match percentages.
七项输入
The Match is computed from seven inputs you provide when you set up your profile.
Input 1
Investment purpose
Rental income, capital appreciation, family use, golden visa, portfolio diversification, or flip and resale.
Input 2
Risk tolerance
Conservative, moderate, or aggressive.
Input 3
Horizon
Short, mid, long, or legacy.
Input 4
Budget
Minimum and maximum ticket size.
Input 5
Preferred bedrooms
Studio through to four bedrooms and above.
Input 6
Financing preference
Cash only, open to mortgage, or payment plan only.
Input 7
Handover timeline
How soon you need the property delivered.
原型
When you complete the profile, you are mapped to an investor archetype. Examples include Balanced Yield Investor and Capital Growth Hunter. Each archetype has a characteristic weighting across the seven Oliva Score dimensions. Your Match is a weighted projection of the project's dimension scores onto your archetype.
The archetype is descriptive, not prescriptive. It is a summary of the answers you gave, so we can explain in plain language why a project fits well or poorly. You can change your archetype at any time by updating your profile.
为何 Match 受限
The Oliva Score is public because it is about the project. The Preference Match is gated behind a free account because it is about you. We cannot personalize without knowing your inputs. Creating an account takes under a minute and does not require a credit card.
隐私
Your profile is stored against your account. It is not sold, not shared with developers, and not used for advertising targeting. You can update or delete it at any time from your profile page. If you delete your account, the profile is removed with it.
更新节奏
Your Match recalculates when you change your preferences. It also recalculates when new market data moves a project's underlying dimension scores, because the Match is a function of those scores. You do not need to take any action. The next time you view the project, the Match reflects the latest state.
我们如何计算回报情景
Each project shows Conservative, Base, and Optimistic return scenarios derived from data about the project's specific community, not a platform-wide default. We build these figures in two steps.
Primary source. We read the Dubai Land Department official sale price index for the community and look at the last 36 months of monthly history. The compound annual growth rate of that index gives us the Base case. Conservative and Optimistic sit one standard deviation below and above Base, using the volatility of the index across the same 36-month window.
Fallback source. When the DLD sale index lacks depth for a newer area, we run a linear regression on individual DLD transaction records for that area, requiring a minimum of 60 observations in the 36-month window before we trust the fit. The regression gives us the same Conservative, Base, and Optimistic triplet.
Sanity bounds. Base appreciation is capped to a realistic Dubai range so edge-case areas cannot produce absurd numbers. Exit values, rental income, and mortgage math are computed on top of the scenario appreciation rate using the same discounted cash flow logic for every project, so differences between projects come from differences in the inputs, not differences in the math.
When we show Limited data. If neither the sale index nor the transaction regression has enough depth to produce a confident fit, we omit the scenarios and flag the section as limited data rather than print a fabricated number.
我们如何推导风险
The risks on the project page are not a generic checklist. Each risk we show is emitted only when the data driving it is present, and its High, Medium, or Low level is set by concrete thresholds on that data. Projects with different profiles get different risk lists.
Delivery risk
Driven by the developer's actual on-time delivery rate, computed from the DLD projects and units registries. A delivery rate above 90% across ten or more prior projects is Low. Seventy-five to ninety percent is Medium. Below seventy-five percent is High. A developer with fewer than two prior delivered projects is flagged Unknown rather than defaulted to High.
Market timing risk
Driven by the 36-month price volatility of the community, computed from the DLD sale index. Areas sitting in the bottom tercile of Dubai volatility are Low. Middle tercile is Medium. Top tercile is High. The description quotes the actual standard deviation so investors can judge for themselves.
Liquidity risk
Driven by transaction velocity in the community, taken from DLD records, combined with the project's own liquidity score. Areas with high transaction volume and short median time on market are Low. Thin secondary markets are High. We cite the actual transaction count and median days on market when the data is available.
Oversupply risk
Driven by the ratio of expected new supply in the next twenty-four months to the area's observed absorption pace. The supply pipeline comes from RERA project filings. Ratios above 1.2 are flagged. The description quotes the actual upcoming-units count and the absorption pace so the investor can reason about the gap.
Macro risk
Driven by the UAE Central Bank policy rate and prevailing mortgage rate, read from the central bank's feed. We flag the risk only when the current mortgage rate sits above 5.5% or inflation is elevated, and we quote the actual prevailing rate in the description.
Regulatory risk
Driven by the project's RERA registration record and the developer's current license status, read from the DLD license registry. Expiring licenses, lapsed escrow status, and disciplinary actions in the last five years each flag the risk. The description names the specific issue rather than emitting a generic warning.
当我们展示有限数据时
Every section of the Investment Analysis carries a confidence label. High confidence means every input behind the section is fresh, comes from a primary authoritative source, and clears the sample size bar. Medium confidence means at least one input is stale or comes from a fallback source. Limited data means we could not produce a number that met our own bar, so we did not produce one.
A Limited data label is a feature of the engine, not a bug. Newer communities have short transaction histories. Newer developers have fewer completed projects. We could fill those gaps with a platform-wide average, but that would mean showing a number that was not actually computed from the project in front of you. That is the behavior we are explicitly engineering against.
Our pledge. We never show a derived number that was not actually computed from the project's own data. If a section would require us to fabricate an input to produce a clean output, we omit the section and show the Limited data label instead. When new data arrives, the section fills in automatically on the next view.
Want every project scored on this methodology?
Oliva runs this 6-dimension, 97-metric scoring engine on 1,000+ Dubai off-plan projects, refreshed nightly from Dubai Land Department records. See your top matches free, or open the full 97-metric breakdown on Pro.
免责声明
- methodology.disclaimer.notAdvice The Oliva Score and the Preference Match are analytical tools, not buy, sell, or hold recommendations. Always conduct your own due diligence and consult qualified advisors before making investment decisions.
- methodology.disclaimer.dataLatency Government transaction data typically has a lag of one to two weeks. Macro indicators are updated as the underlying source releases them.
- methodology.disclaimer.newAreas Areas with very few historical transactions carry lower confidence. In those cases the Score is shown with a data confidence indicator so you know how much evidence supports the number.
- methodology.disclaimer.modelVersioning When we improve the scoring model, we version it. Historical scores on the platform always reflect the engine version active at the time they were computed.
- methodology.disclaimer.offMarket Private deals, bulk purchases, and developer buybacks may not appear in DLD data and can skew area-level statistics.
Last updated
Related Pages