Two Directors, One Treatment: Where Will Their Decisions Meet?

A treatment from a directing duo often contains a sentence like this one: "We will co-direct, with A leading performance and B leading animation." It is accurate. It also buys nothing. It tells the reader there are two of you and roughly where each of you stands, which is a long way from telling them what happens on the day the two of you want different things.

The other common failure runs in the opposite direction — two paragraphs of personal sensibility stapled together, a performance philosophy in one and a drawing philosophy in the other, with no sentence that could only have been written by both of you. A reader can tell. One version publishes an org chart. The other publishes a marriage. Neither explains the film.

What earns the collective voice is narrower and more useful than either: a treatment becomes credible at the exact points where one director's decision cannot be finished without the other's. Name those points, and the "we" has done some work. Leave them out, and "we" is a claim about paperwork — true, unfalsifiable, and worth nothing to the person deciding whether to give you the job.

Find the scene in which the strengths depend on each other

Before anything else, settle the proposition. One film, one sentence, agreed. This piece assumes you have already chosen it; combining two directors' separate film ideas into a third that belongs to neither is a different problem, and a much worse one to bring into a treatment.

Then find the passage where the two of you are structurally stuck with each other. The test is a sentence you should be able to finish:

A's work is ______, which only reads because B's ______ lands on the same beat.

If you can finish that sentence in two different ways about two different moments, you probably have your scene. If you can't finish it at all, you have a sequence — two contributions in the same film, joined by a cut, each one legible on its own. A hard cut from live action into a fully animated passage is like this. It is one shared decision (where the cut falls) carrying two independent pieces of work. That's a handoff, not an interface, and dressing it up as an interface in the treatment will read as inflation.

Here is an invented example, and I'll keep it invented all the way through, because the point is the method rather than a production. A 30-second spot. One performer, waiting at a blank wall. She knocks. Nothing answers. She checks her phone, sighs, and turns to go. Behind her, a handle draws itself onto the wall, turns, and a door opens — warm light spilling across her back. Last frame: her shoulder beginning to turn.

Two directors. A leads performance. B leads the drawn door. Which is the scene that depends on both? Not the knocking; A can shoot that alone. Not the drawn door opening onto a drawn room; B can build that alone. The dependent passage is the four seconds where the film asks the audience to understand that the door waited for her to give up. A's work — the specific quality of a person deciding to stop — only reads because B's door arrives on the same beat, and B's arrival only reads because A's turn happens when it does. Neither half survives a change in the other. That's the scene to write about.

Compare shared direction with a divided route

Now take that passage and run it two ways. Same script, same wall, same four seconds.

Jointly. Both directors are in the room for every take. The door doesn't exist yet, so its timing comes from something practical: an assistant calling the cue, or a monitor beside the camera playing the animatic, or a strip of tape at the door's hinge line and someone's hand on it. The two of you watch each take together and change the cue between takes — this time the handle turns a beat later, this time her sigh comes first — and you find out what the four seconds actually are by running it. The cost is attention. For the length of the rehearsal, both of you are watching one take and each of you is half watching for your own department. The rehearsal gets longer, and neither of you is doing anything else.

Divided. A shoots the performance pass first, against a stand-in cue — a clicker, a hand, a mark on the wall. B builds the door in parallel, getting an uninterrupted run at the drawing, the ink weight, the speed of the swing. Then the two passes meet at an agreed timing review. And this is the honest description of the arrangement: division doesn't remove the dependency, it schedules it.

The asymmetry is the part worth writing into a treatment. In this construction, animation timing is revocable up to a point — a handle-turn can be two frames longer while the scene is still being drawn, and then it can't, without redrawing and recompositing. A performance take is only revocable by shooting again: same room, same light, same performer, same wardrobe continuity. So in the divided arrangement, a late timing change falls hardest on the side that owns the least flexible material — in this construction, the performance. How hard it lands depends on when the two passes meet, which is why the timing review is itself a shared decision, and the most consequential one in the whole arrangement. Move it earlier and the division is mostly harmless. Move it to the edit, and you have asked the performer to be wrong.

Most duos don't choose one of these for the whole job. They choose per passage: separate work where the contributions are genuinely separable, then a joint room for the days that contain interfaces. That isn't a fudge. It's the arrangement, and a treatment that states it is more informative than one that declares a permanent philosophy.

Two things to leave out. Don't propose a default tie-breaker — "the performance director has final say on set" is a labor arrangement, not an explanation of a film, and no reader of a treatment can evaluate it. And don't imply that two directors double your capacity. The same scene may well take longer to rehearse jointly. What two directors buy is not more work done; it's more questions held open, and the treatment should be honest about which ones.

Resolve the interface, not every detail by committee

Now the disagreement, because a treatment that cannot show one has not shown anything. The four seconds contain a real one: does the door arrive while she is still watching, or after she has turned?

If the door arrives while she watches, the performer gets an active beat — she notices the ink, and her face carries hope or disbelief before the door opens. The comedy is generosity: the drawn world answers a woman who is still asking. Her performance now has a reaction inside it, and B's timing inherits a constraint from A's face: the handle's turn has to be slow enough for a person to register it, which puts a floor under animation's speed.

If the door arrives after she turns, the audience knows before she does. The comedy is irony, and it's the version that fits the invented proposition — the world answering when you stop asking. But now A's contribution changes character: the turn away must be a real surrender, not a pause, or the arrival reads as a coincidence rather than an answer. And B's timing gains freedom — the door can move as fast as it likes, because nobody on screen has to see it start.

Notice that both routes reach into the other director's material. That's what an interface is, and it's why the decision can't be made by whoever owns the door.

Which decisions stay local? Use one criterion: a choice is local if changing it doesn't change what the audience knows, or when they know it, relative to the performer. By that test, A owns which pocket the phone goes into, how long the sigh lasts, whether the knock is a joke or a plea. B owns the ink's line weight, the drawn room visible through the open door, the palette of the light. None of those touch the audience's information state, and neither director needs permission for any of them. The cue frame does. The light that lands on her back does — it's a real light on a real person, and it belongs to neither director alone. The final frame does, because it's her shoulder and B's light in the same image.

And the average is the worst answer available. "The door starts to open just as she turns" sounds like a compromise and plays like an accident: the audience loses its advantage over her and she loses her attention to the door, so nobody in the film is doing anything. Simultaneity isn't inherently wrong — plenty of scenes cut on a mutual flinch — but choosing it out of indecision hands the film an information structure that neither director chose, and the result usually reads as a startle rather than a joke.

So how do you resolve it? Not with a rule. Either the proposition decides it — in this case it does, which is why the proposition is worth writing down first — or you test both. In the joint room, that's cheap: run the take with the cue early, run it with the cue late, watch both on the monitor, and the answer arrives as evidence rather than as negotiation. In the divided route you can still test it, but only by building a joint moment inside the division, which is precisely the thing you were dividing to avoid. That's the trade, stated plainly.

Write the collective passage from agreed decisions

Now the prose. Here is the kind of paragraph a treatment gets when the two of you haven't had the conversation:

We're a directing duo with complementary backgrounds in live action and animation. We believe in craft, warmth and surprise, and we'll bring a shared sensibility to every frame. Working with our trusted team, we'll create a seamless blend of drawn and photographic images that feels like one world.

Everything in it is true of any duo that has ever existed, which is the problem. There isn't a decision in it to agree or disagree with.

Here is the passage for the invented spot, in the invented duo's voice, written after the choice above was made.

The wall. Forty feet of it, blank, in flat morning light. Our performer waits in front of it. She knocks. Nothing answers, so she checks her phone, sighs, and turns to go.

The handle arrives behind her — one line of ink, then a shape, then a turn. The door swings inward and the light through it falls across her back. She stops. The last frame is her shoulder beginning to turn.

What the spot is arguing. The drawn world answers when you stop asking. Her surrender and the door's arrival share a beat; neither gets its own shot.

Still open. One thing only, and not the version. The door arrives after she turns; the proposition settled that. What we haven't fixed is how long the gap runs between her turn and the handle's first line — two frames or twenty. We'll settle it in the room, and the test is that the last thing on screen reads as an answer rather than as a coincidence. The door moves at the rehearsal; the criterion doesn't.

Four things are doing work there. The proposition is one sentence in the film's terms. The middle section describes the beat where the contributions meet, and names the interface — a shared beat — without turning into a schedule. The open question is written as a test with a criterion, not as a promise or a committee. And individual expertise shows up as a verb rather than a title: the surrender is performance, the ink is animation, and the reader can see where each of you is strong without being handed two manifestos.

Where names help is in the internal document, not the client one. A scene note that says "the cue is B's number; A's take is downstream of it" tells an assistant director exactly what to protect on the day. In a client-facing treatment, the credit line is irrelevant — the reader is buying the film, and the same dependency can be described without attributing it to a person. Neither version needs a hierarchy, and neither should imply one that hasn't been agreed.

Now the limitation, stated once and meant. The spot above is invented. The two directors are invented, the rehearsal is invented, and the disagreement is invented because I needed a disagreement that could be explained in a paragraph. Nothing here reports how a real partnership works, and a credit line reading "directed by two people" documents nothing about how those two people decided anything — public credits are the org chart again, in smaller type. The method proposed here is a hypothesis about what a treatment should be able to explain.

The check that would test it is specific. Read a first-person duo account in which the pair names a moment where one of them changed something the other had already settled — not their philosophy, their actual moment. Read, with the makers' permission, a treatment or scene note where an interface is visible in the writing. Until then, treat the arrangement above as a proposal with a shape, and take it to a directing practitioner to be argued with rather than to a client as a method.

None of that weakens the last thing a treatment has to do: the ending, which is where a duo treatment usually goes soft. Close on one scene, name what each director owns outright, and name the interfaces where neither of you can finish alone. In the invented spot: the surrender is A's, the arrival is B's, the cue frame and the light on her back and the final shoulder turn are neither. A treatment that can say those three things has shown a reader where the two of you will meet. A treatment that only says "we" has shown them that you are both attending.

Frequently asked questions

What makes a two-director treatment more than an org chart?

It must name the points where one director's decision cannot be finished without the other's. A line saying one leads performance and the other leads animation is accurate but buys nothing. Test for a sentence like 'A's work is blank, which only reads because B's blank lands on the same beat.' If you can finish that in two ways about two moments, you likely have the scene; if not, you have a handoff or sequence, not an interface.

In the invented spot, which passage depends on both directors?

The four seconds where the door waits for the performer to give up. A's surrender only reads because B's drawn door arrives on the same beat, and B's arrival only reads because A's turn happens when it does. The knocking can be shot by A alone, and the drawn door opening onto a drawn room can be built by B alone. Those are separable; the waiting beat is not.

How do joint and divided direction compare on that interface?

Jointly, both directors are in the room for every take and the cue is decided between takes; you learn which version is funny by testing it, but rehearsal is slower and attention is split. Divided, the performance pass is shot first against a stand-in cue while animation is built in parallel, then the passes meet at an agreed timing review. Division schedules the dependency rather than removing it. A late timing change falls hardest on the less flexible material—in this construction, the performance.

Should the treatment include a default tie-breaker or claim that two directors double capacity?

No. A default tie-breaker like 'the performance director has final say on set' is a labor arrangement, not an explanation of a film. Don't imply two directors double capacity; the same scene may take longer to rehearse jointly. What two directors buy is more questions held open, and the treatment should be honest about which ones. Most duos choose per passage, with separate work where contributions are separable and a joint room for interfaces.

How should an open question be written in a duo treatment?

Write it as a test with a criterion, not as a promise or a committee. In the invented spot, the proposition settles that the door arrives after she turns; what remains open is how long the gap runs between her turn and the handle's first line—two frames or twenty. It will be settled in the room, and the test is that the last thing on screen reads as an answer rather than a coincidence. Individual expertise should show up as a verb, not a title.

Further reading

• Write a Shared Director’s Statement Without Pretending Two Makers Think Alike — Once you have named the interfaces, this shows how a shared statement keeps each maker's contribution attributable without inventing a common history or smoothing over what remains undecided.
• Run a Director–Writer Session Around the Scene You Haven’t Solved — When one beat still will not resolve, this session method puts two filmable versions beside it and leaves you with a decision or a named owner.
• Pitch a Series With Multiple Directors Without Promising Identical Episodes — Useful if your treatment claims who decides what; it separates a creative promise about the work from an authority claim that needs someone's agreement first.