Skip to content

Choose a Treatment Writer by Their Judgment, Not Just Their Samples’ Polish

Advertising

Choose a Treatment Writer by Their Judgment, Not Just Their Samples’ Polish

Choose a treatment writer by how they explain and develop film decisions: what they were given, what they contributed, and why the writing took the form it did. Polish is a poor tiebreaker. Two decks can arrive equally clean, equally well-set, using the same three words about restraint, and only one of them was written by someone who can think a film through with you.

This piece gives you a set of questions to ask in a conversation, organized so each answer tells you something the sample alone can’t. I haven’t interviewed treatment writers or the directors who hire them, and I’m not going to describe what the industry reliably does. Treat what follows as a method you can test in the room. The example in the middle is invented, and I’ll say so plainly when we get there.

Start with the collaboration your film needs

Different jobs hide behind the same job title. Before you compare anyone, name which difficulty you’re actually buying help with.

  • You can say it in a room but not on a page. You have a spoken proposal — half-formed, out of order, full of gestures — and you need someone who can hold the mess and hand it back as an ordered piece of writing that still sounds like your film.
  • A transition doesn’t earn itself. The second beat doesn’t set up the third, and you need someone who will question it rather than decorate it.
  • The notes contradict each other. Yours, the agency’s, the client’s, the editor’s. You need them turned into one choice that somebody can stand behind and defend.

Those call for different strengths. A writer who is excellent at ordering your spoken idea may be flat-footed when it comes time to argue with an agency note, and the reverse is just as common. Decide which one this job needs, and let that decide what you’re listening for.

Availability and commercial terms are a separate conversation: who can start in March, what the budget covers, who owns the result, whether they’ll sign your NDA. All of it matters. None of it is evidence about judgment. Settle the creative question first, so the money conversation happens against a clear brief instead of a vague hope.

One more thing before you open a single PDF. A sample can be genuinely accomplished without showing the kind of judgment you’re trying to assess — if it was written from a brief that had already solved the film, or if the director did the thinking and the writer did the typing. That isn’t a flaw in the writer. It’s missing evidence. Don’t penalize someone for not supplying what you never asked for. Ask now.

Ask whose decisions you are looking at

Use a sample the writer is authorized to discuss. That condition comes first, and it shapes everything after it.

Then work backward through the artifact in this order:

  1. What was the brief? The supplied idea, the mandated beats, the constraints, the length.
  2. What did the director bring? The one line, the reference, the thing they were stubborn about.
  3. What were you responsible for? The order, the images, the language, the argument — name it.
  4. Who else worked on it? A designer, an editor, a producer, a copywriter, a photographer.

That fourth question matters more than it sounds. A strong treatment often has four hands on it. If you can’t establish who did what, you aren’t comparing writers. You’re comparing artifacts.

Ask for a bounded explanation rather than the archive. “Could you talk me through one paragraph?” is a good request. “Send me the brief, the client’s emails and the unreleased cut” is not, and you don’t need it. The reasoning can be described out loud. What you want is ten minutes of talk about one page.

If a writer can’t discuss a project — a confidential brand, an unreleased campaign, a client who keeps things close — ask about a different authorized example, or about a film you both know. “I can’t show you that work” is not the same sentence as “I didn’t write it,” and treating it as though it were will cost you good candidates.

Last, a small piece of hygiene: seeing a sample is not permission to reuse it. If you want to reference something, ask, and be specific about what you mean by reference. Writers remember who borrowed without asking, even when they say nothing at the time.

Follow one passage back to its problem

Here’s the invented part. No real writer, client, brand or film is involved, and I’m fixing the facts now so the discussion that follows can be checked against them.

The project: a 30-second film for Ashgrove, a fictional brand, for a fictional refillable deodorant called Mirren. The brief asked for the product to be visible by 0:12, the refill action shown once in full, an on-screen legal line at the end, and no dialogue. Cutdowns were mentioned but not specified — “lengths to follow.” The director supplied one line: “It’s about the half-second of checking yourself before you leave. The click finishes the thought.”

The writer is a candidate. The paragraph below is the one you’ve both agreed to discuss.

It’s 6:40 and the room is still dark. The radiator ticks twice and settles. She stands at the mirror with one sleeve pushed up, checking the same square inch of shoulder she checks every morning, and the check takes less than a second. We stay on her face, not the glass. The case comes into frame — she doesn’t look at it. She fits the refill, the click lands, and she’s through the door before the sound has finished.

Conversation A, limited to tone. The director says the restraint is beautiful, the radiator is a lovely touch, can the writer bring that register to their brand. The writer says yes, that’s the register they work in. Everyone is comfortable. The director has learned that the writer likes quiet films and can say so on request. That’s all that has been learned.

Conversation B, also invented, same paragraph. Six questions, each with a reason to ask it.

“What were you given?” You can’t tell a solved problem from a hard one without the brief. This writer received a one-line idea, three mandated beats and no cutdown lengths. That last gap matters in a minute.

“What did you change? What did you leave?” The change is where the work is. The writer’s first draft opened on the case — the product legible immediately, exactly what the mandate asks for. They cut that and led with the check instead, because the director’s idea wasn’t about the object. It was about the decision, and the click only reads as punctuation if something precedes it. In the approved version the case enters around 0:05 and the click lands around 0:14.

“What did you leave out on purpose?” This separates a decision from a limitation. They cut a commute shot and a second character. Both were choices; neither was an oversight.

“Whose work is this?” The director supplied the half-second idea. The client supplied the mandate. The writer built the passage and its order. The closing pack-shot line — the one you’d notice if you read the whole treatment — was written by the agency’s copywriter.

“What did you get wrong first?” The abandoned opening. It’s the closest thing to a record of thinking that a finished deck will ever give you.

“What would you ask me now?” This is the question that produced a beat in the film. Once the case moved back to 0:05, the 0:12 mandate stopped being automatic — a case can be on screen before 0:12 without its mark ever being legible. So the writer asked what “product visible” actually meant: the case in someone’s hand, or the mark? The answer came back: the mark. So the case turns toward camera once at about 0:07, and that glance is doing the mandated work. It exists because the writer asked instead of assuming.

Follow the paragraph itself too: why does the radiator arrive before the mirror? Which question does “the check takes less than a second” answer? What got cut between the mirror and the door?

A writer who answers “because the check has to feel unremarkable before the click can feel like punctuation” is showing you a decision about your film’s shape. A writer who answers “I liked the echo” is showing you a preference. Both are honest. Only one is what you’re hiring.

Watch for fluency that isn’t judgment. Restraint, elevated, tension, we wanted it to breathe — those phrases can be applied to almost anything, including work the speaker didn’t write. And notice that a convincing answer can include an admitted limit. This writer says the cutdown isn’t solved and they’d want lengths before promising anything. That’s not weakness. That’s the same instinct that produced the 0:07 beat, pointed at their own work.

But it leaves something open. Those 0:07 and 0:14 marks belong to the 30-second film. A six-second version has to land the same action in a fraction of the space, and neither the sample nor the brief shows whether this writer can do that.

Discuss uncertainty, not just the finished answer

Take a different fictional situation. Same candidate, a new project, and the director revises her idea out loud: actually, start on the hands, cut the room entirely. Meanwhile there’s a written note from the agency, sent last week, saying the room must be established before anything else, because the client wants the domestic setting legible early.

Ask the candidate what they’d do in the next ten minutes.

The silent reconciler writes one version that seems to satisfy both and says nothing about the collision. It’s the most common answer, and it’s the most expensive, because the contradiction doesn’t disappear — it just moves downstream to the client review or the shoot floor, where fixing it costs more.

The deferential answer is whatever you want, you’re the director. This feels collaborative and isn’t. It hands the director back both the idea and the mistake she was hiring away from.

The useful answer names the conflict out loud: those two notes disagree; which is newer, and who owns the call? Then it drafts the smallest thing that makes the disagreement inspectable — two openings, three lines each — and states what changes if the room goes. Lose the room and the film loses its setup for “domestic,” so something else has to carry it. Now the director can decide quickly, because the choice is visible instead of buried.

That last part is worth saying twice. Drafting two small openings isn’t hedging and it isn’t invented certainty. Invented certainty is writing a single confident passage that pretends two conflicting requirements already agree, when they don’t.

Collaboration also doesn’t mean agreeing with every sentence a director offers. A writer who says yes to everything isn’t collaborating; they’re transcribing. What you want is someone who takes the revision seriously and then tells you what it costs. Equally, a writer who declares your note wrong without testing it isn’t being rigorous — they’re being difficult. The move you’re listening for is the one that makes the conflict cheap to resolve.

Use a paid trial only for a remaining question

Most of the time, an attributable sample discussion plus a good conversation about competing notes is enough to decide. The trial is for what’s left over.

In the Ashgrove example, one thing is still open: whether the writer can land the same action in a 0:06 cutdown, and whether a click can carry the moment at all if the placement runs muted. That’s a specific uncertainty. If you can’t write your remaining uncertainty as one sentence, you don’t need a trial — you need another conversation.

When you do commission a trial, keep it small and shaped like this. Authorized material only: your own brief, or the writer’s own sample. A short deliverable: say 150 words for the short version, plus the three questions the writer would need answered before writing the full thing. Both halves matter — the passage tests the craft, the questions test the judgment. Time-box it to half a day of their time, not a week.

Agree the use in writing before anything starts: internal evaluation, not submitted to the client, not used in production without a separate agreement, the writer keeps their work, and if you want to use it you negotiate separately. And pay for it, at their rate, for the time.

What to avoid is turning selection into a bake-off. Four candidates, four production-ready treatments of your actual film, is a pitch with a different name, and it costs four people a week of work. If you genuinely must see two writers, keep the task small and pay both, and understand that you’re buying evidence rather than material.

One boundary: whether payment is required and who owns the result are contract and local-law questions, and this is a commissioning recommendation, not a rule and not legal advice. If ownership matters to your project, get proper advice on it.

The note you write before deciding

Three lines, filled in before you choose anyone:

  • Judgment demonstrated: reordered the passage so the click could work as punctuation, then checked which mandate the reorder pressed against instead of assuming.
  • Uncertainty that remains: how the same action lands in a six-second cutdown, and whether a click survives a muted autoplay.
  • Next step agreed: paid half-day; one alternate opening and the questions behind it; internal use only.

If those three lines are full, you’re no longer choosing between decks. You’re choosing between people who have each shown you how they think when the film hasn’t been solved yet — which is where you’ll spend most of the job.

Frequently asked questions

What should you assess in a treatment writer beyond polished samples?

Assess how they explain and develop film decisions: what they were given, what they contributed, and why the writing took the form it did. Polish is a poor tiebreaker. Different jobs hide behind the same title: ordering a spoken idea, questioning a transition that does not earn itself, or turning contradictory notes into one choice someone can stand behind. Availability and commercial terms matter but are not evidence about judgment.

Why can an accomplished sample fail to show judgment?

A sample can be genuinely accomplished without showing the judgment you want to assess if it was written from a brief that had already solved the film, or if the director did the thinking and the writer did the typing. That is not a flaw in the writer; it is missing evidence. Use a sample the writer is authorized to discuss, then ask what the brief was, what the director brought, what the writer was responsible for, and who else worked on it. If you cannot establish who did what, you are comparing artifacts, not writers.

What does the invented Ashgrove example show about a writer's judgment?

The writer received a one-line idea, three mandated beats, and no cutdown lengths. Their first draft opened on the case, but they cut that and led with the check because the director's idea was about the decision, and the click only reads as punctuation if something precedes it. They also asked what product visible actually meant—the case in someone's hand or the mark. The answer, the mark, produced a case turn toward camera around 0:07, doing the mandated work despite the 0:12 requirement. The writer also admitted the cutdown was not solved and would want lengths before promising anything, which the article treats as a limit and not weakness.

How should a writer handle conflicting director and agency notes?

The useful answer names the conflict out loud: those two notes disagree, which is newer, and who owns the call. Then it drafts the smallest thing that makes the disagreement inspectable—two openings, three lines each—and states what changes if the room goes. The silent reconciler writes one version that seems to satisfy both and says nothing, moving the contradiction downstream. The deferential answer hands the director back both the idea and the mistake. Drafting two small openings is not hedging; invented certainty is one confident passage pretending conflicting requirements already agree.

When is a paid trial appropriate, and what should it include?

Use a paid trial only for a remaining specific uncertainty. If you cannot write your remaining uncertainty as one sentence, you likely need another conversation. Keep it small: authorized material only, a short deliverable such as 150 words for the short version plus the three questions the writer would need answered before writing the full thing, time-boxed to half a day. Agree the use in writing before starting—internal evaluation, not submitted to client, not used in production without a separate agreement, writer keeps their work, and payment at their rate. Avoid turning selection into a bake-off. Whether payment is required and who owns the result are contract and local-law questions; the article's recommendation is not legal advice.

More in Advertising Browse all articles