Clear inputs. Transparent rankings.
Seven factors, one configurable score
Popularity 30%; rating 25%; review strength 15%; recent growth 10%; India relevance 10%; freshness 5%; editorial assessment 5%. Scores are calculated in a scheduled job, never per page visit.
How normalization works
Ratings use a 0–5 scale. Popularity, growth, India relevance, freshness and editorial scores use documented 0–100 inputs. Review strength uses log(1 + reviews) divided by log(1 + the largest review count in the category); it is zero when all counts are zero. Weighted normalized values produce a score out of 100. Equal scores use app slug order as a stable tie-break.
What a score does—and does not—mean
Scores reflect the recorded inputs and editorial assessment, not an objective guarantee of quality. Editors must document sources, collection dates and the rationale for subjective inputs. Missing inputs do not become invented observations.
Preview, verification and publication
Demo data is illustrative and physically marked in our database. It is never automatically verified. Editors replace samples with sourced manual records, explicitly verify them, approve the category and publish a complete ranking snapshot. Factual app pages can be reviewed independently of ranking pages.
Updates and movement
The configured refresh runs daily at 02:00 IST, subject to hosting availability. Displayed dates reflect completed snapshots. Movement compares the latest snapshot with the preceding successful snapshot; NEW means no prior position. Failed refreshes preserve the previous snapshot.
Sponsorship and selection
We select apps relevant to each category and review submissions. Submitting or sponsoring an app never guarantees an organic position. Sponsored placements are labelled and cannot change ranking scores.
Data limitations
The MVP uses manual data, not automated store scraping or paid market intelligence. Trending claims require verified growth data. Platform availability, pricing and features may change; consult the official source.