Review a Commercial Treatment for the Film It Actually Promises

Nobody approves the wrong film on purpose.

It happens a page at a time. The story page is moving. The performance note says something precise and true about the actor. The reference board is the best thing anyone has seen all week. Every page passes on its own terms, and by the time you reach sound, you are no longer asking whether the sound belongs to this film. You are asking whether the sound is good. It is.

A treatment is a proposal for one film, written in several languages at once: the brief restated, the scene told as prose, a performance note, a camera plan, an edit shape, a sound idea, and a set of references with captions. Each section is easy to approve alone. The review that catches a real problem is narrower and slower than a read-through. Pick one moment the film depends on. Follow that moment into every part of the document that describes it. Check whether they all ask the reader to imagine the same thing happening.

Start with the assignment and the film you say you are making

Read the brief before you read the treatment, and read it as a separate document. The brief contains the things you cannot argue with: the product, the length, the mandatory end frame, whatever the client has already signed. The treatment contains a proposal about how to satisfy them. Those are different kinds of statement, and mixing them is how a review turns into a rewrite of somebody’s taste.

Write down the interpretation in one sentence before you look at anything else. Not the plot. The interpretation: what this film is about, and what the audience is supposed to feel at the end. If you cannot write it in one sentence, that is already the finding, and it is more important than anything on page six.

Cardboard Spaceship’s guide to reading a director’s treatment before awarding the job (checked 8 September 2026) makes two moves worth keeping. It separates the source material from the execution, and it asks whether the proposed film is understandable. It is first-party guidance from a production company; it does not measure whether treatments that read well get chosen more often. Neither does this review. But the distinction is the right one to start from, because almost every treatment problem is a source-versus-execution problem wearing a costume.

Then pin the version. Treatments get revised in shared documents with three people typing. Name the draft you are reviewing, so that “I fixed that” and “it’s still on page nine” can be settled by looking rather than remembering.

A constructed example. Nothing below is a real treatment, a real client, or an observed audience response. The document is stipulated so the repairs can be followed.

Bracken Paper sells a correspondence set: folded cards, envelopes, sealing wax. The brief asks for a 30-second film, music only, no dialogue, no voiceover. Two jobs are mandatory — the card and its handwriting must be on screen long enough to be read, and the film ends on the packshot with the brand line. A 15-second and a 6-second cutdown will follow.

The director’s chosen interpretation is that the spot is not about the invitation. It is about the second before someone decides to offer it. Restraint is the settled approach; the pleasure of the film is watching somebody almost not do something.

Follow one moment through every section that describes it

The moment this film depends on is the handover, beat 3. Everything else in the document either sets it up or pays it off. Here is what each section says, and what each asks the viewer to imagine.

Some of this needs no comment. Beat 3 and the performance note agree, and they agree in the way that matters, which is about action and timing rather than adjectives.

Beat 3. Mina sets the envelope on the cloth, between the bread and the salt. She does not push it toward him. Her hand stays on it for one count, then goes back to her lap. Daniel is still talking. Mina looks at her water glass. He turns back, sees the envelope, and reaches for it. Only then does she look up.

Performance. Mina is tentative throughout. She is not selling the invitation. No smile until it has been picked up, and then only a small one. The hesitation is the scene; we are watching someone decide whether to offer something at all.

Then there is REF 03, and this is where the review has to read carefully.

REF 03 — “Grand Lobby” (hotel film). A woman crosses a marble lobby and presents a box to a man at the far table with a flourish, total assurance, the room noticing. Warm, tactile palette, long lens, shallow depth. This is the table texture for beat 3.

The caption describes two different things and calls them one. The reference’s dominant action is a confident, public presentation with a flourish. Its texture is warm, tactile, long-lensed, and shallow. Texture is portable. Action is not, or not without changing the scene. The caption claims to be borrowing the texture and then names the action as the reason.

Read every caption against what actually dominates the reference, not against the adjectives glued to it. A reference can be useful for its grade and its lens and still be the wrong model for what people do in the shot. The word “assurance” also appears in the tone strip and in the caption that cites REF 03 later in the document — three passages, one vocabulary, all three asking for what the performance note forbids.

That is a contradiction, not a counterpoint. It is worth saying plainly, because the difference matters:

• Contradiction: two sections ask for incompatible things and the document never says why.
• Counterpoint: two sections ask for different things and the treatment explains what the difference is doing.

Is the sound cue a contradiction or a deliberate contrast?

Sound. A brass-led cue with real swagger. It arrives on beat 3 and peaks as the envelope lands.

Taken alone, this looks like the fourth passage wanting confidence. But a confident musical flourish under a hesitant gesture is one of the oldest jokes in film, and a good one. The music can do the bragging the character will not. The film sounds sure of her even though she is not sure of herself. That is not a contradiction; it is the clearest statement of affection in the whole document.

The reason to keep it is not that it is funny. It is that the cue only works because the performance is restrained. Underneath an assured performance, the same cue stops being a joke and becomes a description. So the cue and the interpretation are the same decision wearing two hats. If the film ever flips to the confident route, the cue has to be rewritten, because the contrast that generated it will have disappeared.

That is the test to run on any apparent clash of image, performance, and sound. Ask what the difference is meant to do, then ask whether the treatment says so. Here it doesn’t say so — the sound section describes the cue and never names its job. One sentence would fix it, something like: the cue is not Mina’s confidence, it is the film’s; she does not know it is playing. Without that sentence, the actor, the editor, and whoever handles the temp music will each resolve the tone differently, and the film will arrive in the edit as a negotiation.

A counterpoint can be kept whenever the document explains it and nothing nearby depends on the thing it contradicts.

Fix the page, or send the question back

Restraint is settled here, so most of this is repair work, and repair work is mostly following the word you removed into the places it was cited.

The tone strip goes first, because it is the line everyone reads before they read anything. “Warm, assured, celebratory” was doing real damage; “assured” is exactly the quality the performance note prohibits. Something closer to warm, unhurried, and on her side would describe the film that beat 3 describes.

REF 03 does not have to be thrown out. A reference kept with a negative instruction is a normal, useful thing: borrow the palette and the long lens for beat 3, and do not borrow the entrance — in the reference, the offer is a performance; ours is not. The caption then describes what is actually being taken, and the reader knows which parts of the reference are furniture and which are weather.

REF 07 gets the same treatment and one more job. It is a good reference for a hand sealing an envelope, which belongs in the packshot insert rather than the table scene. “Same assurance as REF 03” comes out. And wherever the camera section cites REF 07 as a model for Mina’s hands, that citation moves to the insert, because brisk, sure fingers on that table is a different film.

The camera section’s “presentation” becomes “the handover.” One word, borrowed from the vocabulary that was removed, is enough to put a small performance back into a scene that is about not performing.

Then there is the hole rather than the contradiction. The document never says whether Daniel knows what the envelope is before he reaches for it. If he does know, his reach is a rescue, and Mina’s hesitation has a witness, which changes the scene. If he does not know, his reach is casual, and her decision is a lonelier and better thing. The treatment needs one clause in beat 3 to say which, because the reader cannot follow the moment without it.

Now change the stipulation. Suppose restraint is not settled. Suppose the director is still deciding between two films, and the reference caption is honest about wanting the second one.

Then the caption is not the error. Beat 3 and the performance note become the stale parts, and a caption repair would be worse than useless, because it would bury a live creative decision under tidier wording.

• The hesitation film. Camera waits with her. The envelope is on screen before she commits. Nothing in the frame congratulates anybody. The reward is Daniel reaching first. The swagger cue is a joke on her behalf; she never hears it.
• The sending film. The pleasure is in the giving. She slides it across and watches him open it, and the room noticing is the point rather than the error. The camera moves with her instead of waiting. The cue matches her, and there is no joke left in it.

Both are coherent. Neither is a wording problem. The choice belongs to whoever owns the creative decision — the director, and whoever approved the interpretation in the first place. If the agency or the client already signed off on the hesitation film, the other film needs a conversation, not a caption, and the person assembling the document should not quietly pick a side while performing a word count.

What must not happen is an average. Language such as “restrained but confident,” “quietly explosive,” “intimate yet epic” looks like a resolution and isn’t. It hands the actor two incompatible instructions, and it hands the editor two incompatible scenes. Each will solve it alone, and the disagreement will come back in post, where it is more expensive and less fun to settle.

Read the revised film using only the document

Put aside everything you know about the project and read the action again as a reader who has only this text.

A Sunday lunch is ending. A guest brings an envelope up from her lap and sets it on the cloth between the bread and the salt. She does not push it toward him. Her hand stays on it for one count and then leaves. The host finishes his sentence, turns back, sees it, and reaches. She looks up only then, and when he opens it and looks for her, she is already looking away. The card is legible in her handwriting. The music has been bragging about her the entire time, and she has no idea. The packshot holds, then goes.

That is one film. Nothing in the walkthrough requires an explanation the document does not supply, which is the actual test of a repaired treatment.

Then check that repairs have not left old endings behind. A corrected caption is a claim about a scene, so look at every page that cited it. If REF 07 has moved to the insert, does anything still refer to it as the model for the table? If the push-in now ends at the handover rather than the presentation, does the edit section still cut on a beat that no longer exists? If the final shot rewards the room noticing, the assured film is still in there somewhere, and it will win, because endings are what people remember.

Close the review with two short lists.

Resolved: the tone strip no longer asks for assurance; REF 03 specifies palette and lens and explicitly excludes the entrance; REF 07 has been rehomed and uncoupled from REF 03; the camera section says handover; beat 3 states whether Daniel knows what he is reaching for; the sound section names the cue’s job, which is affection for her that she never hears.

Open, and needing a decision rather than an edit: where the legible card lives in the 6-second cut, since the brief requires legibility and the cut currently keeps beat 3 and the packshot; and whether the restrained interpretation has been agreed with anyone outside the treatment.

Everything else belongs to other reviews. This one says nothing about whether the wax seal reads on a phone, whether the location is available, whether referencing a hotel film requires anything, whether the budget covers a long lens, or whether the client will buy it. Cross-section agreement proves that a document describes one film, and nothing else. Feasibility, permissions, and approval each need their own evidence, and they do not get any of it from a story that finally agrees with itself.

Frequently asked questions

What is a practical first move when reviewing a commercial treatment?

Pick one moment the film depends on, then follow that moment into every section that describes it. Check whether all sections ask the reader to imagine the same thing happening. Each section can be easy to approve alone, so the review should be narrower and slower than a read-through.

How should I separate the brief from the treatment?

Read the brief as a separate document. It contains what cannot be argued with: product, length, mandatory end frame, signed requirements. The treatment proposes how to satisfy them. Write the interpretation in one sentence before looking further: what the film is about and what the audience should feel at the end. If you cannot, that is already the finding.

How do I handle a reference whose caption seems to conflict with the film's core moment?

Read every caption against what actually dominates the reference, not the adjectives attached to it. A reference can be useful for grade and lens while being the wrong model for action. Texture is portable; action is not, or not without changing the scene. A reference can be kept with a negative instruction: borrow the palette and lens, but do not borrow the entrance.

When is a clash between sound, image, and performance a contradiction rather than a counterpoint?

A contradiction is two sections asking for incompatible things with no explanation. A counterpoint is two sections asking for different things and the treatment explaining what the difference is doing. A brass-led cue under a hesitant gesture can be a deliberate joke, but the treatment should name the cue's job; otherwise actor, editor, and temp music will each resolve the tone differently.

What if the treatment's core interpretation is not settled?

Then caption repairs may bury a live creative decision. A hesitation film and a sending film are both coherent but different: in one the camera waits with her, in the other the pleasure is in the giving and the room noticing. The choice belongs to whoever owns the creative decision. Do not average them with phrases like 'restrained but confident,' which hand the actor and editor incompatible instructions.

Further reading

• Camera, Editing, and Sound Can Tell Different Versions of the Same Event — Extends the cross-section check to a second invented scene, showing how camera, edit and sound build compatible or incompatible accounts of the same twelve seconds.
• Hand Over an Unfinished Treatment Without Making the Next Writer Start Blind — Once your review leaves open questions, read this to hand the document on with each unfinished decision labelled and its owner named.