Write a Commercial Treatment When Important Parts of the Brief Are Missing
Write a Commercial Treatment When Important Parts of the Brief Are Missing
A brief with a hole in it invites a particular kind of cheating: writing the film you can write. The treatment comes back clean, the client reads it, and two weeks later a fact arrives that turns the spine of the thing over. Nobody lied. The document just behaved as though the hole weren't there.
The alternative isn't to stop work until the brief is complete. It's to sort the gaps by what their answers could overturn.
For any unknown, ask what changes under each plausible answer. Three outcomes, and only three:
- Nothing that matters changes. Continue the work, and say in the document that it holds either way.
- Two different films come out. Fork narrowly, label the fork around its consequences, and name who can collapse it to one.
- The answer would make the treatment assert something nobody has established. Hold that work in plain sight.
The measure is rework, not gravity. A question that sounds enormous — which of two kitchens — can be a color swap in a document and a redressed corner on the day. A percentage on a deck slide can be the whole film. Judge the unknown by the creative decisions standing on it, not by how serious the missing subject sounds.
An unknown is only as large as what it can overturn
The first read of the brief has already found the gap. What happens now isn't a second checklist; it's a routing decision. Four questions get you most of the way.
Does it survive to the shoot? A gap you can answer after the footage is in the can is a different animal from one that has to be answered before a set is built, a location is held, or a performer's day is booked. Move the answer's deadline to the point where the decision stops being free. That's the real deadline, and it's usually earlier than the brief's.
How much already-decided material rests on it? If the tag line, the character's behavior, the world, and the tone all bend when the answer changes, the gap is structural. If only an insert and a prop list move, it's surface.
Is it a fact, a preference, or a constraint? Preferences can be defaulted — pick one, mark it, move on. Facts and constraints can't be resolved by taste, no matter how confidently the treatment is written. A placement plan, a clearance window, a cutdown count the post schedule can't absorb: those are constraints, and they belong to whoever holds them.
Is the answer close? A question that resolves tomorrow costs almost nothing to wait for. One that resolves in three weeks costs the calendar. That asymmetry decides more routes than the creative stakes do.
Then ask the question that keeps a gap from becoming a shoot-day improvisation: who can answer this, and by when? An unknown without an owner isn't an unknown. It's a decision that will get made by whoever is closest to the problem at 7 a.m. with the truck running.
Keep working where both answers agree
Name the material that is identical under every plausible answer, and write it fully — unhedged. In a commercial that's usually more than a writer expects: the emotional relationship, the setting, the talent's behavior, the fixed product shot, the tag, the tone. That's real work, and none of it is wasted.
Then record the gap beside the work it would change, with the limit stated precisely:
This section holds whether the kitchen reads family-forward or pared back. The refrigerator insert does not.
One line, next to the thing it governs. Not a footnote at the back, not a caveat that quietly drains the writing. Hedge the gap, not the film.
There's a distinction here worth being strict about. Rewriting two sentences in a document costs two sentences. Committing a set build, a location fee, a performer's day, or a sound design pass built around a composed motif costs money and calendar, and those commitments belong to the client, not to the treatment. A treatment doesn't become reversible by being written in pencil. The document is cheap; the thing it describes may not be.
When the answers make different films, fork small and label it
When a gap changes the story rather than the surface, you're in the second outcome from the top of this piece: two different films come out. Inside that outcome there are three routes, and it's worth comparing them honestly:
Wait. Costs calendar, saves rework. Right when the dependent work is large or the answer is genuinely imminent. Waiting is not passive; it's the cheapest route on this list when the conditions hold.
Fork. Costs a little drafting, protects the calendar, and puts the choice in front of someone who can make it. Right when the divergence is small — a beat, an ending, a line.
Narrow. Continue only the shared spine and hand over less film. Right when the uncertain work turns out to be most of the piece.
These three belong to the middle outcome, and only to it. They aren't a second closed list to route gaps with: a new unknown still goes through the three outcomes at the top, and wait, fork and narrow come up only once the answer lands in that middle case. The third outcome — a claim nobody has established — isn't on this list at all, and the next section says why it's held instead of routed.
A fork has rules. Fork at the point of divergence, not at the top — share everything above the punchline, so the reader sees one film and then two exits. Write the exits at the smallest scale that actually shows the difference: a few lines each, with the consequence of each, not a menu of aesthetics. And frame the question around what changes, not around what's better. No client can answer "which ending is funnier." Most can answer this:
If the social run starts muted, the chime gag doesn't land and we shoot the steam tell instead — one insert and a second mix pass, and we lose the beat of silence.
One caution. Don't write three complete speculative treatments because the brief was thin. Three finished documents transfer the directing job to a reader with less time and less context than you, and they invite a fourth round. A fork that grows past half a page is usually a sign the underlying choice is bigger than a fork — which is a reason to ask, or to wait, not a reason to write more.
Label assumptions where they can't be missed. Assumption: placement is sound-on. Then the consequence. Not a footnote — a line in the body, in the argument's path.
Some gaps aren't creative — hold them
Separate creative uncertainty from an unsupported factual claim. A creative choice can be defaulted and revised later. A performance claim, a certification, a comparison to a named competitor, a delivery promise: those aren't unresolved because the brief is thin. They're unresolved because nobody has established them, and a treatment is not where they get established.
A conditional note doesn't fix it. "Subject to substantiation" changes the paperwork, not what the audience takes away. If the dramatic payoff of a scene is a number, the film asserts the number. Whether it can be asserted, with what qualifier, in which markets, is not a treatment question — and possibly not only a marketing question. Claim verification and legal review are outside what a writer can settle here, and pretending otherwise is worse than an unfinished section.
The same restraint applies to resources. A location that hasn't been scoped, a performer who hasn't been asked, a cutdown count the post schedule can't absorb — a treatment that proposes them is committing other people to things they haven't agreed to. The proposal may look decisive. It is actually a promise someone else has to keep.
Holding work has a real cost: the document looks less finished. That's the price of a document that's honest, and it's usually cheaper than the alternative. Preserve the shared work so the file doesn't read as an apology. A treatment with a written spine and one honestly held section is a stronger artifact than a complete-looking one built on a guess.
Say what each route costs, and who has to answer
Whatever route you take, state the trade plainly: what the delay costs, what the limited parallel development buys, what a narrower proposal gives up. And name a role, not a department. "Brand lead" gives someone something to do. "The client" gives no one anything.
Then say what changes after the answer arrives — which sentence gets deleted, which insert gets added, which section gets written. The reader should be able to see their own answer landing in the document.
Keep it readable. A treatment that reads like a change log is a treatment nobody finishes, and the assumptions will be skimmed past by exactly the person whose decision matters. Few assumptions, clearly marked, in the path of the film's argument.
One boundary worth knowing. The April 2020 AICP National Guidelines — Live Action Production, in the commercial production bidding section, on printed page 1, tie a production proposal to clear specifications from the contracting client. That's trade association guidance from a specific date, written for a bidding context. It isn't law, it says nothing about treatments, and it doesn't validate any method described here. What it supports is narrow and useful: the proposal is meant to answer specifications. Where a specification doesn't exist yet, a proposal that appears to answer it is misleading, whether or not anyone intended it that way. That's the whole reason to mark the gap in the document.
Trade guidance can frame a conversation. It doesn't settle budget authority, compensation, permissions, or legal obligations, and a treatment shouldn't pretend to.
One appliance, three gaps
The example below is invented to show the routing. No client has run it, and the answers aren't the point — the point is that three gaps get three different routes for the same reason: different rework.
The brief: a fictional countertop steam oven, the Halden S-2. Thirty seconds plus cutdowns, six-week window. Fixed material: the audience is weeknight cooks who've stopped trusting their oven; the tone is dry, not slapstick; the tag reads "No plan required."; the door-closing product shot is required. Three things the brief doesn't say.
Gap one: does the kitchen read lived-in or pared back? The deck's mood pages contain both a cluttered kitchen with a school bag on a chair and a bare concrete counter with one bowl. Nobody noticed they conflict.
Owner question: to the brand lead — is this film positioning the S-2 as a family machine or a design object? That's a positioning call, not a taste call. The director can write to it but can't settle it, and anything left standing in its place has to be labelled as exactly that: a provisional reading the draft holds to, which only the brand lead converts.
Shared under either answer: the opening beat through the door closing — hands, the oven door, the parent not watching it, the dry tone, the required door-closing shot — plus the tag at the end. The door closes around nine seconds.
Depends on it: a refrigerator insert, a background performer at the table, the prop-loaded counter corner, art department scope.
Route: continue — the first outcome. Write the spine in full. Add one labelled line carrying a provisional default and its consequence: family-forward keeps the fridge insert and the kid; pared back loses the insert and the film runs two seconds shorter. The brand lead's answer replaces the line. If it flips, the document changes by a sentence. After a set build, it's a redressed corner and a pickup — not a new location. Notice what not to do here: don't let one dressing question grow into a set design document. The gap is real and small.
Gap two: does the primary placement run with sound? The brief lists connected TV and social cutdowns and doesn't say which leads, or whether the social version autoplays muted.
The joke in the current draft: the parent hums two notes while closing the door, and twenty minutes later the oven's timer answers in the same two notes. With sound, that's the turn of the film. Muted, it's a person standing near a counter looking at a timer.
Owner question: to the media lead, because placement is what sets the mute condition. This is close to a delivery constraint dressed as a creative question, and it should be routed like one.
Shared: everything up to the door closing — the same nine seconds described under gap one.
The fork, three lines each. Route A, sound-on leads: the two-note answer, one beat of silence, tag — and a composed two-note motif in the sound budget. Route B, muted-first: the parent's double-take at a curl of steam against the oven glass, the chime kept as a bonus for anyone listening, the silent beat cut and a second mix pass booked.
Route: fork — the second outcome. Under half a page, and the setup survives either way. Waiting costs nothing in rework and everything in calendar. Note the consequence shape: the punchline, a sound budget line, one insert, and a few seconds of running time all move together with one answer.
Gap three: the speed figure. The deck says the S-2 reaches temperature "in a fraction of the time," and one slide carries a percentage with no source. No substantiation, no qualifier, no market list.
The tempting scene writes itself: a split screen with two timers, a countdown insert, the parent glancing at a clock. That scene makes the number the drama. Once it does, the film asserts the claim.
Owner question: to the client's product-marketing or regulatory owner. Not the director, not the agency creative, not the producer.
Held: the countdown insert, the split screen, the figure in any voiceover line.
Survives: most of the film, because the agreed idea is that the machine asks nothing of the parent — the subject is attention, not speed, and the parent's unavailability is the comic engine. That doesn't depend on the number.
There's a limit worth stating out loud. If the client's position is that the figure is the premise rather than a slot the film can do without, then the whole film is held, and that's not a triage problem. It's a conversation about whether there's a film yet. Say so rather than papering it over with a footnote. And don't imply the things the claim would drag in behind it: a demonstration shoot, a comparison, a testing protocol. Those carry cost and possibly exposure, and they aren't the writer's to offer.
Three gaps, three routes. The dressing question got a labelled default and a sentence. The placement question got a two-route fork. The claim got held. Same brief, different logic — and the difference is entirely about what each answer would overturn.
What you send, and the question you end on
The document that comes out of this has a shape: the shared spine written fully; one labeled line with a default and its consequence; a fork of a few lines with the answer that collapses it; a held section naming what's held and who owns it; and a short note up top saying which parts are firm, which are conditional, and what the reader's answer changes. It reads like a film and behaves like a plan. Both properties matter.
End with the question, not with the gap. "Please confirm the kitchen dressing, placement, and substantiation" hands the work back unshaped. Instead:
If social leads muted, we cut the chime gag and shoot the steam tell — the setup, the cast, and the tag are unchanged, we lose the beat of silence, and we add a mix pass. Confirm by Tuesday and the fork collapses to one ending.
That's the shape of a good next question: it names a decision, the consequence of each answer, and the point at which the answer stops being useful.
A brief with a hole in it isn't asking you to guess better. It's asking you to be specific about what you don't know — and to keep making the parts of the film that don't depend on it.
Frequently asked questions
How should a treatment handle a missing brief fact without guessing?
Sort the gap by what its answer could overturn. If nothing that matters changes, continue and say it holds either way. If two different films come out, fork narrowly and label the fork. If the answer would make the treatment assert something nobody has established, hold that work in plain sight. The measure is rework, not how serious the missing subject sounds.
What questions help route an unknown in a brief?
Ask whether the gap survives to the shoot; how much already-decided material rests on it; whether it is a fact, a preference, or a constraint; and whether the answer is close. Then ask who can answer it and by when. An unknown without an owner is a decision that will get made by whoever is closest to the problem when production is running.
When should a treatment fork, and how large should the fork be?
Fork when the gap changes the story rather than the surface. Fork at the point of divergence, share everything above it, and write the exits at the smallest scale that shows the difference—a few lines each, with consequences. Frame the question around what changes, not what's better. Don't write three complete speculative treatments; a fork that grows past half a page usually means the underlying choice is bigger.
Which gaps should be held rather than solved creatively?
Hold unsupported factual claims—performance claims, certifications, comparisons to named competitors, delivery promises—because a treatment is not where they get established. A conditional note like 'subject to substantiation' changes the paperwork, not what the audience takes away. Also hold resource commitments to locations, performers, or cutdown counts that other people haven't agreed to. State what is held and who owns it.
How should a treatment with missing brief parts end?
End with the question, not with the gap. Name a decision, the consequence of each answer, and the point at which the answer stops being useful. For example: if social leads muted, cut the chime gag and shoot the steam tell; the setup, cast, and tag are unchanged, but you lose a beat of silence and add a mix pass. Confirm by Tuesday and the fork collapses to one ending. The shared spine should still be written fully.