BFI Industry Data & Insights

It is the British Film Institute's official statistical account of its own market. The BFI - the UK's lead film body and a National Lottery distributor - publishes the Industry Data & Insights hub as the country's record of film and the wider screen sectors, produced as official national statistics on a scheduled release calendar.

Three product families sit under one name. The weekly UK box office report ranks the top fifteen films released in the UK each weekend alongside all other British releases and newly-released films, with Friday-to-Sunday grosses in pounds sterling, weeks-on-release, cinema counts, site averages and cumulative totals, and an archive running back to 2017. The Statistical Yearbook lands annually as a full volume plus companion data tables, stacking editions back to 2002 including UK Film Council-era volumes, and spanning box office, top films, UK films and talent worldwide, distribution, exhibition, home video and television, audiences, production, certification, public investment, employment and the wider UK film economy. Scheduled official statistics then add quarterly and annual cuts of UK box office, film and high-end television production, and certification, published at a fixed morning hour on announced dates. Research reports, READR evidence reviews and Research and Statistics Fund outputs round out the hub.

One row per film per reporting week, already ranked. The mono block above shows the leading record off a recent weekend chart plus the next two rows down the ranking.

Read the top row as three facts already decided for you. The money arrives in sterling and split three ways - the weekend itself, the per-site average, and the cumulative total - so a holdover analysis never re-divides anything. Trajectory travels on the row: Weeks on release turns a single weekend into a point on a decay curve without joining a second file. Because every row names its distributor and country of origin, share-of-screen analyses by studio and by producing nation are group-bys, not research projects. And a brand-new entry reads differently by design: its week-on-week change arrives empty and its cumulative total equals its opening weekend, which makes new-versus-holdover segmentation a filter rather than a heuristic.

Ten fields define the weekly chart row, every definition verified during the August 2026 research pass. They divide into three jobs:

• Identity: Rank, Film and Country of Origin say what is playing and where it came from, with origin carried natively rather than inferred from cast lists.
• Money: Weekend Gross, Site average and Total Gross to date give the weekend, the intensity and the lifetime figure in pounds sterling - all three typed as currency, none needing a recompute.
• Exhibition: Distributor, % change on last week, Weeks on release and Number of cinemas describe the release machine itself: who put the film there, how hard it held, how long it has run and how wide it plays.

The full dictionary follows in tabular form below; anything not pinned there folds under additional fields on request. The yearbook's tabulated series - distribution, exhibition, home entertainment, audiences, production, certification, employment - ride along as separate structures keyed to edition year when your question needs the market behind the weekend rather than the weekend itself.

• Geography - the United Kingdom as one national market. Every gross is a UK gross and every cinema count is a UK screen count; the yearbook widens the lens once a year to UK films' worldwide performance. For territory-by-territory international grossing outside this frame, pair it with Box Office Mojo.
• Temporal - weekly box office from 2017 to present, roughly fifty-two charts a year; yearbook editions annually from 2002; scheduled quarterly-plus-annual official statistics on top. Launch windows, holiday corridors and franchise curves are all reconstructible at week grain across nearly a decade.
• Granularity - film-level weekly rows in the headline product, roughly 780 ranked rows a year of current weekly data, each carrying its week explicitly so incremental pulls append cleanly instead of re-pulling history; national aggregates for the production, certification and market statistics around it.

Against the catalog - 1,744 datasets averaging 7.81 on the quality rubric - this record's 8/10 rests on verified field documentation and a decade-deep weekly archive. Where it ranks: best movies & entertainment datasets.

API, files, or your warehouse. Daily, weekly, or hourly.

Six jobs the weekly chart settles outright:

1. Distribution strategy - read holdover curves and site-average decay week by week before pricing P&A commitments against comparable openings.
2. Exhibitor programming and screen allocation - widen, hold or cull on the number-of-cinemas versus site-average trade-off while the decision still matters; see competitor tracking.
3. Studio and distributor share analysis - rank the releasing companies themselves off the Distributor column, week over week and year over year.
4. Investment and M&A diligence - anchor cinema-chain and content-slate valuations in official national figures; see market sizing.
5. Citation-grade verification - any quoted number resolves to a named film, weekend and rank; see citation grade research.
6. Policy and public-funding evaluation - the certification and production series exist precisely to measure whether intervention moved the UK screen economy, and the yearbook supplies the long baseline.

Each job maps to a persona below, and the sample validates whichever one you came for.

Journalists, Academics & Students lead fit: official national statistics cite cleanly, resolve to named weekends and survive review, with a 2002-forward yearbook shelf behind any claim about British film - see movies & entertainment data for journalists, academics and students at /for/journalists-academics. Market Researchers & Consultants build UK market-sizing and share work on the one box-office series produced by the industry's own statistical authority - see movies & entertainment data for market researchers. Competitive Intelligence & Product Teams track rivals' slates and distributor momentum in near-real time; see movies & entertainment data for competitive intel product teams.

Persona fit has edges worth naming. Data Scientists find no per-user signal here - this measures markets, not raters, so modelling jobs stay anchored on rating corpora such as GroupLens MovieLens Datasets. Developers & Builders get no artwork or metadata payload to enrich a UI with. This is the ledger of what happened at the box office, not the library of what the films are.

Three notes worth having upfront.

First, the weekly chart is a ranked cut, not a census: top fifteen plus British releases plus new openers, which is effectively the whole commercial story but not literally every screen in the country. Long-tail repertory and event cinema sit outside it.

Second, grain changes between products: film-week rows in the headline chart, national aggregates everywhere else. State which layer your question needs before scoping a feed.

Third, money is nominal: grosses arrive in pounds sterling of their own weekend, with no deflation applied. A 2019-versus-2025 comparison needs a price index applied downstream, and admissions are not implied by revenue - ticket-price drift alone can manufacture a fake boom. Then get a sample of this dataset cut to your titles and weeks.

The ten-column core pins every weekly row. Beyond it the hub extends along four axes, delivered as extra structures when a sample or feed calls for them:

• Yearbook data-table series - distribution, exhibition, home video and television, audiences, production, certification, public investment and employment tables from editions back to 2002, delivered as their own frames keyed to edition year.
• Scheduled official-statistics series - quarterly and annual UK box office, film and high-end television production, and certification figures from the fixed release calendar.
• Research and market-intelligence outputs - topic studies, READR evidence reviews and Research and Statistics Fund findings, mapped to the series they contextualise.
• Archive depth - earlier weekly charts beyond the current window, delivered on request so a backfill starts where your analysis does rather than where a default export stops.

Ask at sample request and these ride alongside the same week keys as the core rows.

Because the raw artifact underneath is a weekly spreadsheet built for one person reading one weekend, and every interesting question wants years of weekends, every distributor and the yearbook's annual context in one frame. Datadory hands over the corpus itself: the ten-field dictionary kept verified, week keys normalised so time-series code never parses a filename again, currency typed as currency, and the annual yearbook tables joined onto the weekly spine when your question spans both - landed on the cadence your pipeline runs, with real rows in hand before any commitment.

Browse the rest of the slice on the movies & entertainment data hub, see how an official market measurement squares against a modeling corpus in the BFI Industry Data & Insights vs Kaggle The Movies Dataset comparison, or read the weekend box office chart glossary entry for the vocabulary underneath the numbers.