One Treatment Collaborator or Several Specialists?
One Treatment Collaborator or Several Specialists?
By the time you are choosing between arrangements, you already know the treatment needs writing, research, and design. That question is settled. What is not settled is which of those jobs can move independently and which ones keep pulling on each other. The arrangement that serves the work is the one whose structure matches that dependency pattern — and the pattern is usually smaller and more specific than the staffing conversation assumes.
So the decision is not one person versus three. It is: where does judgment have to stay connected, where does specialist depth actually change the result, and who owns the integration when a change lands. Answer those three and the headcount follows, sometimes in a mixture you did not expect. Answer them carelessly and you can hire a strong team whose polished pieces quietly argue for different films.
The comparison below runs one invented project through one late change under three arrangements. Everything about the project is fictional, including the transit system, the stations, and the copy. It is a scaffold for the method, not a record of anyone's job.
Map the decisions that travel between words and images
Start with a few consequential decisions, not a chart of who does what. The chart is an output of this work, not a substitute for it.
Picture the treatment page as three layers that already exist on it: the reference, the written explanation, and the page treatment itself. Each layer carries its own kind of decision, and most of them can proceed on their own. Choosing which morph to point at as a reference is a judgment about visual precedent, and it rarely needs to wait on anyone outside the design conversation. Confirming a factual claim needs research and does not depend on how the frames are drawn. Setting the easing and timing of a move is craft, and it can be solved without reopening the argument.
The trouble lives in the small number of decisions that keep returning to another discipline. A good way to find them is to take one proposed action and trace it all the way across the page.
In the fictional spot, the action is a single continuous move. A rider's paper ticket drops into a fare slot, the stub curls, and the curl becomes the arc of a rail line on a map, and the map resolves into a train window. It is a pretty piece of choreography. It is also not only choreography. The map is making a claim about what a fare buys, so the move depends on a research finding. The line that sits beside the frames — in the invented version, "One fare. Every line on this map." — is written against that same claim. And the reference the designer chose was selected because a window is a place your eye can rest; the reference is doing work for the destination the move lands on.
Notice what that gives you. There is one genuine loop here, and it runs through three decisions: what the move lands on, what the fare actually covers, and what the copy promises. Everything else on the page is freer than it looks.
That is typical. A treatment has a lot of parallel work and a few loaded joints. You are not deciding how to staff the parallel work. You already know how to do that. You are deciding who holds the joints.
Test the integrated route against actual skill and capacity
One collaborator holding the loop means one person carries the transition's choreography, the fare claim, and the copy that rests on it. That is a real advantage, and it is easy to overprice.
The advantage is continuity of judgment. When the landing point of the move changes, the person who wrote the copy already knows the copy was written against the old landing point. They do not have to be told. They feel the strain before anyone reports it. That kind of continuity is hard to buy back later, once the pieces are separate.
But continuity is only worth something if the person can actually make or coordinate the judgments the loop requires. Which means you inspect relevant work rather than the title. A collaborator who has written and designed static pages beautifully may never have resolved type legibility inside a fast move. That is a bounded skill, and its absence is not a character flaw. It just means the loop still needs a contribution it will not get from this person.
The second thing to test is capacity, and it is the failure mode people skip. When every question in the loop returns to one person, the loop has a single thread. If that thread is occupied — on another project, on a question you have not answered, on a fact that has not come back — the whole loop waits. Note what the waiting actually looks like: it is not a slightly slower treatment. It is a decision you cannot make because the person who understands the dependencies is not available to say what breaks if you change them. That is a bottleneck, and it is a structural property, not a bad week.
There is also a subtler cost. Continuity can slide into sameness. A collaborator with a reliable move will reach for it, and the move may substitute for the decision. When one person holds everything, the treatment can become very coherent and slightly inevitable — solved the way they always solve it. That is not a reason to avoid the integrated route. It is a reason to ask what the loop looked like the last time it was genuinely difficult.
Test the specialist route against integration demands
The team route starts from a fair premise: writing, research, and design each reward depth, and one person rarely has all three at a high level. Assign the contributions and that premise pays off. The researcher can confirm the fare coverage with a precision the generalist would have guessed at. The designer can build frames the generalist would have settled for.
The premise only holds if the integration is owned by someone. So make that person visible before you start, not after the first conflict. The test is concrete: follow one change through the handoffs and ask three questions. Who notices it? Who revises the affected work? Who checks that the treatment still describes one film?
Run the late change — the director decides the move should land on the departure board rather than the window — through a specialist team that never named an integrator.
The note reaches the designer, because a landing point looks like a design decision, and it is one. The designer rebuilds the frames: the curl now resolves into a departure board with four rows of destinations and times. Filling those rows, the designer picks plausible commuter destinations, and one of them is Fort Marne. It reads well in the row and sounds like a place the line would serve.
The writer is not in that conversation. The line "One fare. Every line on this map." stays as it is, and it was written when the map was the last thing on screen.
The researcher is not asked, or is asked only whether the station names sound real, which they do. Nobody reconnects the board to the earlier finding that a single fare covers the core zone and Fort Marne requires a separate fare.
Every component is finished. The frames are clean, the copy is intact, the names are plausible. And the treatment now argues that one fare takes you to Fort Marne, which the research had already ruled out, through a station the researcher would have flagged if the board had ever reached them. The contradiction is not in any piece. It is in the seams, and no one owns the seams.
That is worth stating precisely, because it is the argument the team route tends to lose. This failure is not a capability gap. Everyone was capable. It is an authority gap. A named integrator on this same team — the same three people, one of them additionally responsible for the whole — would very likely have caught it, because the catch is just somebody holding the loop. Specialist depth is real, and it is not the thing that was missing.
More contributors also do not remove your own decisions. The board versus the window is a directorial call. The team can inform it; the team cannot make it for you. And when two finished components disagree, somebody still has to overrule one of them. That somebody is usually you, whether or not you planned to spend the morning on it.
Compare the same late change under both arrangements
The comparison only tells you something if it is fair. Do not run your best integrated collaborator against a hypothetical disorganized team, and do not run a careful team against a hypothetical overwhelmed generalist. Use the same project, the same late change, the same four destinations, and the same Fare-coverage finding. Then look at four things: decision continuity, specialist depth, waiting points, and repeated work.
Integrated collaborator. The change lands. The one person knows the map was built to the fare-coverage finding, so they know immediately that a board with Fort Marne on it reopens a closed question. They revise the landing point and reconsider the copy in the same pass, before the two drift apart. Decision continuity is high. Waiting point: everything, until that person is free and the fare-coverage question is refreshed. Capability question: whether the board's type will hold up in motion is a bounded skill the collaborator may or may not have.
Specialist team, no named integrator. The change lands on the designer. Frames are revised well. Depth on the visual side is high. The copy and the research are not pulled into the change, and the treatment fragments into three correct components that describe slightly different films. Waiting points are short and scattered, which disguises the problem, because each person is waiting only a little, and nobody is waiting on the contradiction. Repeated work appears late, when someone finally reads the assembled treatment and finds the board contradicting the line above it.
Integrated lead with a bounded specialist. One person carries the loop — copy, research, and the transition — and a second contributor is brought in for exactly one task: making the board's type legible at the speed of the move. When the change lands, the lead catches the copy implication and the fare-coverage implication together, then scopes the type problem into a brief: these destinations, this duration, this screen. The specialist returns type that fits; the lead puts it in place and checks that the treatment still describes one film. The risk here is at the seam between them. The lead has to recognize that motion typography is a genuine specialist problem and scope it, and if the destinations change again, the specialist is on the critical path a second time.
Set the three side by side and the diagnoses separate cleanly. The solo route's weakness is a bottleneck and possibly a bounded capability gap. The team route's weakness is unclear authority, not thin talent. The mixed route's strength is that it keeps the loop in one pair of hands while buying depth only where depth changes the result — and its weakness is that it depends on the lead's judgment about where the boundary lies. A lead who cannot see the boundary gets the mixed route's handoffs without the mixed route's depth.
Assign a working route and a way to reopen it
Choose against the dependencies you mapped and the actual people you have. For the invented project, the dependency pattern — one small loop running through landing point, fare claim, and copy, plus one bounded specialist need in motion type — points at the mixed arrangement. An integrated lead holds the loop and integrates changes; a specialist contributes the one task the lead cannot do as well. That is a decision about this project, not a verdict on headcount.
Write down three things plainly, because they are what fails first.
Who integrates. One named person, not a role and not a meeting. This person owns the seams. Their job is not to do all the work; it is to notice when two finished pieces stop agreeing and to route the question to whoever can resolve it.
Where open questions return. The fare-coverage question belongs to research, and its answer has to reach whoever sets the landing point before the frames are rebuilt. Name the return path, or the question disappears into the gap between two conversations.
Which assumption would reopen the choice. For the mixed route, the load-bearing assumption is that the lead can see where their own skill ends. If the lead's motion-type work turns out to be strong, the specialist is unnecessary and the arrangement collapses back to one person with no loss. If the project turns out to need two genuinely separate research domains — not one fare question, but two — then the lead's attention stops being the integrator and starts being the bottleneck, and a second contributor alongside a named integrator becomes the better structure.
Keep two things out of this decision. Availability and commissioning terms are real constraints, but they are not creative suitability; a collaborator who is free next week is not thereby the right fit, and one who is busy is not thereby wrong. And nothing here argues that specialist teams or single collaborators are generally better, faster, or cheaper. There is no universal advantage in either direction. There is only the question of where judgment has to stay connected, who is holding it, and what would have to be true for that to change.
For this project, that means one integrator and one bounded specialist, with the fare-coverage question routed back to research every time the transition's landing point moves. The assumption that keeps it standing is the lead's accurate sense of their own limits. Watch that assumption. When it stops holding, the arrangement should change with it.
Frequently asked questions
How do you decide whether one collaborator or several specialists should work on a treatment?
Do not start with one person versus three. Map the decisions that travel between words and images: where judgment has to stay connected, where specialist depth actually changes the result, and who owns integration when a change lands. Trace one proposed action across the page. Most page work can proceed independently; a small number of decisions keep returning to another discipline, and those loaded joints determine the arrangement.
In the invented ticket-to-train-window example, what is the real dependency loop?
The loop runs through three decisions: what the move lands on, what the fare actually covers, and what the copy promises. The map makes a claim about what a fare buys, so the move depends on a research finding. The line beside the frames is written against that same claim. The chosen reference is doing work for the destination the move lands on. Everything else on the page is freer than it looks.
What are the advantages and risks of one integrated collaborator?
The advantage is continuity of judgment. When the landing point of a move changes, the person who wrote the copy already knows the copy was written against the old landing point. The risks are capacity and capability: if every question in the loop returns to one person, the loop has a single thread and the whole treatment waits when that person is unavailable or blocked. Continuity can also slide into sameness, where a reliable move substitutes for the decision. Inspect relevant work and test actual capacity, not title.
What goes wrong on a specialist team with no named integrator?
The article's invented late change—the move lands on a departure board rather than a window—shows the failure. The designer rebuilds the frames and picks a plausible destination, Fort Marne. The writer is not in that conversation, so the copy stays as written when the map was the last thing on screen. The researcher is not asked or is asked only whether station names sound real. No one reconnects the board to the earlier finding that a single fare covers the core zone and Fort Marne requires a separate fare. Every component is finished, but the treatment now argues something the research ruled out. The contradiction is in the seams, and no one owns the seams. It is an authority gap, not a capability gap.
What should be written down before assigning the treatment arrangement?
Write down who integrates—one named person who owns the seams; where open questions return—for the invented project, the fare-coverage question belongs to research and must reach whoever sets the landing point before frames are rebuilt; and which assumption would reopen the choice—for the mixed route, the load-bearing assumption is that the lead can see where their own skill ends. Keep availability and commissioning terms out of the creative suitability decision. The article says there is no universal advantage for specialist teams or single collaborators.