Launch timing against filed supply: when the register says go

Developers time launches by festival calendars, funding cycles and courage, and then meet, on launch weekend, every competitor who timed theirs identically. The register offers something no previous market cycle had: the competition's forward pipeline, filed with dates attached, quarters before their hoardings rise. Timing against that curve is the closest thing to legal insider knowledge this industry offers, and it is public.

The forward curve nobody plots#

Every new registration in your catchment is a competitor announcing, under oath, what they will build and roughly when: the registration's promised completion date, its unit count where one is filed, and a quarterly progress record that refines both. Stack your catchment's registrations by promised delivery and you hold the forward supply curve, the chart that says which quarters face three rival completions and which face none. Most developers have never drawn it for their own corridor, which is why launch weekends keep hosting reunions.

It is drawable because the date is filed on essentially everything. Across the 19,650 active registrations in Maharashtra, every single one carries a promised completion date, which is rare for this register, where most fields are filled by a minority. Here is what they say when you stack them.

2027 is the crowded year: 5,585 active registrations promise completion in it, 28 percent of the whole active register, and the curve thins fast after 2028. That is the statewide shape, and it is not the one you launch against. Your catchment's own curve is what matters, and it will look nothing like this one: a corridor whose competitors all promise 2028 is a different bet from one whose pipeline empties after next year.

An analogy: the airline's fare calendar#

Airlines price the same seat differently by departure date because they can see booking curves and competitor capacity by day. A developer with the register's forward curve holds half that instrument: the same tower, launched two quarters apart, meets different competing capacity and therefore different pricing power. Nobody at the airline says "we always launch fares at Diwali"; the calendar answers to the capacity curve, and the capacity curve is, in Maharashtra real estate, a public document. The half the airline has and you do not is the booking curve, which is the next section's whole problem.

The demand line, and why it is the hard half#

The supply side of this exercise is filed. The demand side is not, and a method that pretends otherwise will draw a confident chart out of nothing.

MahaRERA publishes a project's current state and keeps no archive of its earlier filings. Open a project today and you get the sold count as it stands; open it next quarter and you get next quarter's. There is no stored history, so a trended absorption series over the last eight quarters cannot be reconstructed from the portal by anybody, however patient, and a consultant who shows you one has modelled it from transaction data rather than read it off the register. Two things are available instead, and both are real.

The cross-section, today. Read sold against sanctioned for every catchment project beside its age on the register: many projects of different vintages, photographed on one day. That is not velocity, and must never be labelled as it, but it answers the question a launch decision actually asks, which is how sold a two-year-old project gets in this lane. The method post sets out the full version, and the Pune district read is the worked example: across 48 pincodes there, standing unsold stock runs from 32 to 69 percent of sanctioned units, and the lane you are in matters far more than the year you pick.

The series, if you started recording. Movement is also rarer than a quarterly chart implies. Over a seven-week window we observed, 1.6 percent of projects filed a changed sold count, so a catchment of twenty projects may show two or three moves in a quarter. A catchment is the right unit precisely because single projects are too quiet to read.

Drawing and reading the curve#

The go signal, and the false ones#

The register's green light is compound: a thin forward curve in your delivery window, standing unsold stock at the light end of the district spread, and few recent registrations nearby. One component alone misleads. A thin curve in a dead catchment is thin for a reason, which is why the catchment test precedes every timing exercise; low standing stock with nothing registering is a small market rather than a hungry one; and a busy catchment whose curve is empty after next year may simply be one where everyone has already finished.

The false signals are older than the register: festival-season conventions, which every rival reads from the same calendar; funding pressure wearing strategy's clothes; and the rearview trap, launching into last year's hot corridor exactly as its wave lands. The curve exists to overrule all three.

Iqbal reads his corridor's curve#

Iqbal's third Thane project (illustrative, as ever) launched on exactly that reading. His curve showed two rival registrations promising completion in successive quarters and nothing registered behind them, while the catchment's standing unsold share sat near the light end of the district's range. He registered early, filed impeccably through approvals, and opened bookings into a corridor whose quiet his afternoon exercise had seen coming.

Note what he did not get from the register, because it is the part that gets overclaimed. It told him how much competing stock was promised and when; it told him nothing about which product to build, since MahaRERA files no unit configuration, no carpet area per flat and no price, and nothing about how fast the corridor had been selling last year, since the register keeps no history to answer with. Those came from his own sales data and his own recording. The register decided the quarter, not the building.

Where to draw it free, and what a report adds#

The supply half is assemblable by hand at no cost. The free district pages list every registered project in a district with its status and its promised completion date, and each project page carries the units, the construction position and the extension history, so a catchment of a dozen competitors is an afternoon with a spreadsheet. That is also the right way to learn what the filings do and do not say before paying anyone for a summary of them.

What costs money is repetition and memory. The area market report is a flat Rs 2,999 per pincode and assembles the catchment's forward curve, its standing stock and its distress markers against the district cohort, and the Rs 2,999 project analysis does the same around a specific parcel with the two and five kilometre rings drawn from the filed coordinates. Their real advantage over the afternoon version is the archive: the figures are read against filings recorded over time rather than against whatever the portal shows this morning, which is the one thing the portal itself will not keep for you.

The one-line summary#

The competition files its completion dates in advance and every active registration in the state carries one, so stack the pipeline, read standing demand as a cross-section rather than a trend the register cannot supply, launch into the thin quarters and after the waves, and let the festival calendar decorate a decision the curve already made.

Methodology and sources#

• Promised completion dates, project counts and the forward curve: ReraGenie analysis of 55,913 published MahaRERA projects, as updated on 21 September 2026, restricted to the 19,650 currently reading Active and grouped by each project's own filed proposed completion date. All 19,650 carry one.
• Unit totals are summed from the building-level table, which about three quarters of these projects file, so every unit figure above describes filers and undercounts the true pipeline. Project counts are complete.
• The filed date is the CURRENT promise, extensions included, which is why only 1.3 percent of active registrations sit past their own date: a project that slips usually files a new date rather than staying overdue. The curve therefore describes intentions as they stand today and moves when promoters re-file.
• Standing unsold shares and the 48-pincode Pune spread: the same corpus, active registrations filing the building-level unit table.
• Booking movement: ReraGenie trend ledger, 15 observation dates between 31 July and 17 September 2026 across 58,573 projects carrying a filed sold count; 960 projects, 1.6 percent, filed a changed count in that window.
• MahaRERA files no unit configuration, no carpet area per flat and no price, and publishes no archive of earlier filings. Nothing in this method splits the pipeline by product type or derives a historic absorption rate, because the register cannot support either.