This page exists because a number without a source is a rumour, and because the fastest way to check an analysis is to read what it admits it does not know. Everything below is generated from the same data that builds the project pages, so it cannot drift out of date relative to them.
Coverage is uneven. We publish the reach of each source rather than quietly showing you only the projects where everything happens to be available.
| What it is, and what it gives us | Projects reached |
|---|---|
| The developer’s own filing to the regulator Every promoter must file a disclosure before selling: total project cost broken into land, construction and development charges; units built and units booked; the completion date promised and any later revision. It is the strongest provenance available to us, because the developer signed it. | 376 46% |
| The state real-estate register Registration number, issuing authority, promoter entity, and whether the registration has lapsed. We match projects to it by name, and only a confirmed match is allowed to carry a legal claim on a page. Where we hold the developer’s own filing reference, the registration number on the page links straight to it on the regulator’s own site — 386 projects, and rising as more filings match. | 378 46% |
| The town-planning licence rolls Which developer holds a licence for which colony, over how many acres, and whether an occupation certificate has been granted. Also every application the department refused. This is what a developer page is built from. | 114 14% |
| Published prices and configurations Asking price, unit sizes, and the area basis each rate is quoted on, which matters most. A carpet-area rate and a super-area rate are not comparable, and most comparisons you will see elsewhere silently mix them. | 802 98% |
| The resale market What comparable homes in the same building are being asked for now. Asking prices only: Haryana does not publish transaction values, and anyone quoting them to you is estimating. | 113 14% |
| Residents What people living there report, positive and negative, in their own words. We do not average it into a star rating; we show what they said. | 214 26% |
| Position on the map Coordinates for each project, used for distance to a station, to employment centres, and for the commute search. | 800 98% |
Every number on a project page carries one of these. They are not decoration; a sheet containing any modelled line may not be used as a quotation.
Taken directly from a document or a register. Stamp duty at 7% is sourced: it is the statutory rate. A developer's declared project cost is sourced: they filed it.
Estimated by us, with the assumption stated. Where a project has not filed its development charges, we model them — and the page says the figure is modelled and what it assumes.
We do not know. This is a designed answer. A module that cannot answer says so and says why, rather than filling the space with something that reads like an answer.
A low-density flagship compared against the average of its sector reads as "116% above the area" — technically true and completely useless. Instead each home is set against homes quoted on the same area basis, in the same price segment, at the same stage of construction. Where fewer than four such homes exist, we say the comparison set is too thin rather than making one up.
We tested whether future returns were predictable from what we hold, on data held back from the fitting. They were not. Rather than publish a forecast we know fails its own test, we publish history with an uncertainty band and label it history.
It combines nine signals — how demand compares to supply for that specific building, how many towers finish alongside it, where the price sits among peers, ten-year sector growth, distance to a station and to work, and what residents complain about. It is a transparent weighted index, published with the contribution of every signal, and it is deliberately not a trained model: there are no recorded resale outcomes to train one on, so a model would be fitting an invented target.
Filling in a missing value would invent the exact fact you came here for. Where a signal is absent we re-weight across what remains and publish how much of the picture we actually had. Below 55% we do not publish a score at all.
The most important section on this page.
If a figure here is wrong about a project you know well, tell us at hello@trusite.in. Corrections are how the record improves, and we would rather be corrected than confident.