Pulse reads a film’s public signals before release — trailer views, search interest, social volume, booking-page interest, audience polls — converts them into the number of people who intend to buy an opening-weekend ticket, and reconciles that against the screens and seats actually available. The output is a projected collection with an honest range around it.
These are demonstration figures
Raktdhaar, Dilli Se Dubai and Kaali Raat are invented titles. Their signal inputs were written by hand to exercise the model across a wide range, so the numbers below are model output on fictional data. They are not projections for any real release, and nothing here should be read as a forecast for a film in cinemas.
A high-buzz title
Strong trailer numbers, heavy booking interest and warm audience polls twelve days out.
A mid-size release and a quiet one
The same model on thinner signals. Notice how the projected collection and the Buzz Index fall together, and how the range around the collection stays wide when the inputs disagree.
How to read a Pulse figure
- The number is the median of two thousand simulations, not a single guess.
- P10 – P90 is the range those simulations fall in. A wide range means the signals disagree with each other.
- The arrow compares today’s reading with the previous one, so a figure embedded in an article keeps moving as release approaches.
- The source line names the model and carries the timestamp, in IST, of the reading on screen. Every figure we publish is dated.
- Buzz Index is a 0–100 summary of the signals. Ticket intent is people, not rupees. Projected lifetime collection is India nett.
After release
When a film opens, its projection is frozen and the actual result is recorded beside it. Figures for released titles show both, with the gap between them, so the model’s record is visible rather than quietly revised. None of the demonstration titles above has released, so no comparison is shown yet.
Every Pulse figure updates itself once a day at 09:00 IST for as long as the film is tracked.