Pitch a Documentary Before You Know How It Ends
Pitch a Documentary Before You Know How It Ends
You have footage, a subject, and a situation that hasn't resolved. The funder's form asks for a Film Summary, and you have to describe a film whose second half does not exist. So you write the ending you're hoping for — the transfer goes through, the community rallies, the archive survives — and you send it out knowing you can't deliver any of it.
The failure isn't exactly a lie. It's a drift. It happens sentence by sentence, because a proposal reads better when the story sounds finished, and the future tense is so easy to reach for. By the last paragraph you have promised a transformation you have no authority to promise, and the reader has no way to tell which parts you know and which parts you want.
There is a better version of the same document, and it is not weaker. It promises a method of inquiry and attention instead of control over reality. Everything below is about building that document from what you actually have.
Begin with a present situation, not the importance of a subject
"Community media preservation matters" is not a film summary. Neither is "This archive is at a crossroads." Both are claims about significance, and the reader will either accept them for free or not at all. What the reader can't supply on their own is the situation: who, where, what material, what pressure.
A subject becomes a situation when something has to happen and hasn't yet. The difference shows up in a single pair of sentences:
- Preservation is urgent for small radio stations across the country.
- The station has to be out of its building by the end of March, and no one has decided which of its roughly 1,900 tapes come along.
The second sentence is a film. It has a place, a quantity, a deadline, and an open decision. It also has an ending nobody possesses — not the station, not the volunteers, not you.
For the rest of this article I'll work with an invented packet so the decisions are concrete. Everything about it is fictional: a community radio station with a collection of about 1,900 reels and cassettes recorded between 1974 and 1998, mostly airchecks, live in-studio sessions, and meeting tapes. The station is losing its building. The collection has to be gone by a date the organizer has stated. Nobody has decided what gets kept.
Notice that the pressure is not yours. That's useful, and worth saying out loud in the summary, because it tells the reader where the film's momentum comes from. A deadline that belongs to someone else is a real engine; a filmmaker's desire to make a meaningful film is not.
Separate recorded events, reports, and future opportunities
Before you write smooth prose, annotate the material. Four statuses are enough:
Captured. Footage, audio, or images you have. In the invented packet: two cataloging sessions, about nine hours total, filmed since January, plus one 70-minute recorded interview with the organizer.
Reported. Something a participant has told you. The organizer says the building must be empty by the end of March. The lease itself has not been reviewed, so the deadline is a report — evidence that she said it, not evidence of the date.
Established. Facts you or someone on the team verified directly. Two people counted the shelves and boxes in January and arrived at roughly 1,900 items.
Hoped for. The request to film the sorting meetings, made in February. No answer yet. That is a request, not a permission, and the fact that sorting meetings would make a wonderful scene does not move them into the third category.
Some of the most damaging sentences in proposals come from letting those categories slide while drafting. "The building must be empty by March 31" drops the attribution and turns a statement into a fact. "We will film the volunteers dividing the collection" puts a scene in the present tense of your commitment when the access doesn't exist. "The lease requires the tapes to be moved" upgrades a report to a document nobody has read.
This isn't an eccentric worry. A 2025 proposal checklist from the Nonfiction Core Application asks applicants, in the Film Summary, to distinguish what remains unknown from the film's present account. That is one program's guidance for one application cycle, not a universal rule for every deck, and it does not tell you how much uncertainty a particular reader will tolerate. But it does suggest the distinction is being looked for rather than forgiven.
There's a practical reason to care beyond any checklist. Your annotated inventory is the thing you'll consult every time the project changes. If the deadline moves, you need to know whether you're updating a fact or a rumor. If the request to film the sorting meetings comes back as a no, you need to know exactly what the proposal promised before you can rewrite it.
State the live question and the method for following it
The live question is the one your film's attention can actually pursue. "Will the collection survive?" is not that question — it's a coin flip dressed as an inquiry, and no camera makes progress on it. The observable version is narrower: What gets kept, and on what grounds does anyone decide?
That question can advance in a single afternoon. Every item that goes into a keep box or a discard pile answers a piece of it. The film is making progress even when the larger outcome is completely unknown, and that's what makes the proposal credible.
Now the method. "We will see what happens" is not a method. Say where you'll be, what you'll watch, what you'll ask, and what you'll do when someone declines to be filmed. In the invented packet, the method might be three commitments: film the handling — what gets picked up, what gets set aside, how long a tape stays in someone's hand; ask everyone who touches the collection the same question, "What makes a tape worth carrying?"; and keep the camera on the material rather than on any single person's journey.
That last choice matters, and it's often where filmmakers talk themselves into a problem. You do not need a protagonist. A film can follow objects, testimony, or a recurring procedure. Here, the tapes are the through-line and the people rotate past them — which is not only structurally honest for a collection that many people handle, but also sturdier. If one volunteer stops coming, the film doesn't collapse, because no single volunteer was carrying the transformation.
The question and the material have to be genuinely connected. If your stated question is "how do communities decide what memory is worth keeping?" and your footage is a room full of people arguing about shelf space, there's a gap the reader will feel, and it reads as vagueness. Name the observable version of the question — the one the sorting table can answer.
Show how different outcomes would change the film
This is the section that separates a serious proposal from a hopeful one. You are going to describe plausible futures without assigning anyone a motive or a choice, and you're going to be specific about what each one does to the film.
If a partner archive accepts the collection. The tapes move to climate-controlled storage, sorting happens on a schedule, and the film becomes an argument about criteria: what gets kept, by whom, on what reasoning. Strongest of the three for the observational method, because the decisions happen in rooms you may be allowed into.
If the deadline holds and there's no room for everything. Some portion doesn't survive the move — shelving emptied under time pressure because there's nowhere to put it. The film becomes one about triage: what gets saved when saving everything is off the table, and what the phrase "saved" can mean when it covers only part of the shelf.
If the decisions are postponed. The tapes get boxed into a rented storage unit, the deadline passes, the boxes stay sealed. The sorting scenes never happen. That is a different film about deferral — or, if the boxes go behind a locked door and nobody opens them, no film from this premise at all, which you should say plainly rather than promise three equally moving versions.
Notice what none of those paragraphs do. They don't say volunteers will realize the collection's value, or that the station will find the money, or that someone will step forward. They describe conditions, not choices. "If the sort happens, someone has to decide which of 1,900 items stay" is a structural statement about the situation. "The volunteers will decide to keep the live sessions" is a prediction about people who haven't met yet.
Also notice that you're allowed to rank the outcomes. Readers can tell when a proposal pretends every branch of the future would produce a great film, and it makes them trust the rest of the document less, not more. Say which outcome gives you the film you're describing, which gives you a different one, and which would send you back to the drawing board.
Write the summary with uncertainty still visible
The order that works: present material first, live question second, method third, and unresolved conditions wherever they change the reader's understanding of any of the three.
Then do a future-tense audit. Read every sentence with a will in it and ask who controls the verb. If the subject is you and your crew, future tense is a commitment you can keep — we will film the cataloging, we will ask the same question of everyone who handles a tape. If the subject is a participant, an institution, or an event, future tense is a prediction wearing a commitment's clothes.
The reverse failure is worth guarding against too. Uncertainty is not permission to withhold what you know. If you've counted the tapes, give the number. If you've been told a date, give the date and its source. A summary so hedged that it contains no facts isn't careful; it's evasive, and it wastes the credibility you'll need later.
Here is the invented packet's summary with the categories intact:
Our film follows roughly 1,900 reels and cassettes — airchecks, live sessions, and years of station meetings recorded between 1974 and 1998 — that have to leave the building. Volunteers have been cataloging the collection on camera since January, a few hours at a time. According to the organizer who oversees it, the building must be empty by the end of March. No one has decided what gets kept. We have asked to film the sorting meetings, if sorting happens.
We don't know where the collection ends up. What we can film now is the work of deciding: what gets picked up, what gets set aside, what gets argued over at the table. We ask everyone who handles the tapes the same question — what makes a tape worth carrying?
The film could end in a new storage room, in a dumpster, or in a rented unit with the boxes still sealed. Those are three different films. If the boxes go into storage and stay there, the one described here won't exist, and we'd tell you that rather than deliver something else.
Read it against the annotated inventory and every claim holds. The count is established. The recording dates are established. The deadline is attributed. The cataloging is captured. The sorting access is requested. The endings are conditions.
That's the whole job, and it doesn't require a single invented scene. It does require you to live with an open question in your own document for as long as it takes to finish, which is harder than it sounds when the deadline for the application is close and the fictional version would read so much better.
What the summary can promise, and what it can't
A documentary proposal is read as evidence about how you work. A reader who finds a prediction in it doesn't just doubt that sentence; they start scanning for the next one. A reader who finds a clear account of what you have, what you're asking, and what would change your approach learns something more useful: that you'll still be there when the situation moves, and that you'll describe it accurately when it does.
For the invented packet, the present promise is modest and checkable — 1,900 tapes, a stated deadline, nine hours of cataloging footage, one pending request. That's enough to make a decision on. It's also the kind of promise you can keep.
One limitation belongs beside this whole method. The outside support for it here is narrow: a single checklist item from one 2025 funding application, plus editorial reasoning about how proposals get read. It hasn't been tested against a published filmmaker statement from an unfinished project, and the radio station above was invented to demonstrate the categories, not observed. The distinctions themselves — captured, reported, established, hoped for — you can test on your own material this afternoon, which is where the method was always meant to be checked.
Frequently asked questions
What should a documentary pitch say when the ending is unknown?
It should start from a present situation rather than a claim about importance: who, where, what material, and what pressure. Present material first, the live question second, the method third, and unresolved conditions wherever they change the reader’s understanding of those three. The pitch can promise a method of inquiry and attention, not control over reality.
How do the categories captured, reported, established, and hoped for prevent overclaiming?
They separate what you have on camera, what a participant has said, what has been verified directly, and what you have only requested or hope for. A reported deadline should keep its attribution; a hoped-for scene is not permission or access. Letting those categories slide turns a report into a fact or a request into a commitment.
What makes a live question filmable rather than a coin flip?
The filmable version is narrow and observable: for example, what gets kept and on what grounds anyone decides. It can advance in a single afternoon as items go into keep boxes or discard piles. 'Will the collection survive?' is not that question, because no camera makes progress on it.
How should a writer audit future-tense sentences in a summary?
Read every sentence with 'will' and ask who controls the verb. If the subject is you or your crew, future tense is a commitment you can keep. If the subject is a participant, institution, or event, future tense is a prediction wearing a commitment’s clothes. Also avoid the reverse failure: uncertainty is not permission to withhold facts you know.
What limits does the article place on its own method?
The outside support is narrow: a single checklist item from one 2025 funding application plus editorial reasoning. It has not been tested against a published filmmaker statement from an unfinished project, and the radio station example is invented to demonstrate the categories. The distinctions themselves can be tested on one’s own material.