Design a One-Take Commercial Around Attention, Not the Camera Stunt

By pitch.dogFoundation12 min read20 September 2026

Advertising

A one-take commercial is pitched twice, even when the treatment document only shows once. There is the version in the room with the client, which sounds like a camera decision, and then the version that actually earns the money, which is an attention plan. What follows is a way to build the second one, and to hold onto it under pressure.

Say what the cut would cost you

Start with a subtraction. A cut is not a gap in craft; it is a tool. If your idea works just as well cut into six shots, the one-take premise has not earned its place yet. The first question isn't "can this be one shot?" but "what does the viewer lose if I cut?"

Three effects tend to be genuinely continuous:

Accumulation. Pressure builds across a single unbroken interval—a waiting room, a queue, a slow countdown. Every cut releases a little tension, so a cut is a way to breathe. A true uninterrupted take denies the viewer that release. The continuity is doing the emotional work.

An unbroken exchange. Two people negotiating, reconciling, avoiding each other. What matters is that neither gets an exit. A cut between them would imply a moment of privacy or reprieve that the story doesn't want to give them.

A discovery. Something becomes visible through a connected sequence of causes—the parcel that moves from a stranger's hands into the hands of the person waiting for it. The continuity is the argument that this chain of hands is one event.

Notice the wording of the last clause: "one event." Not "immersive." Continuity is not automatically immersive, authentic, calm, or immediate. It only produces the effect your staging and timing give it. A long unbroken shot of nothing reveals nothing faster than a short cut.

So state your effect in a sentence you can defend to a producer: If we cut here, the audience loses the sense that the sender's writing and the recipient's recognition happen in one continuous moment. If you can't name what's lost, you haven't yet found your reason.

Map attention, not territory

A one-take proposals sometimes collapses here. It describes where the camera goes ("drifts down the corridor, past reception, turns left into the loading bay"), which is a map, not a design. A moving camera does not guarantee that anybody notices anything. The job is to hand attention from one point to the next.

In the margin of your treatment, before you write any prose, sketch an attention route. Write down, in order:

The anchor. What holds the viewer's eye at the start: a hand, a face, a sound, a moving object.
The handoff. The single most reliable cue that tells the eye where to go next: a gaze, a gesture, a sound, a movement through frame.
The reveal. When the new focal point is unmistakable.

The handoff is the hinge. A camera move that arrives somewhere without a cue leaves the viewer scanning the frame to figure out what mattered—and the sense of continuity, which was supposed to be the point, now feels like disorientation.

Establish the geometry you need. If the parcel leaves the packing bench and the recipient is two rooms away, the audience needs to understand, early, that these two rooms are adjacent—or the route reads as a series of disconnected plates. Let a character glance at a doorway, or let the human chain itself show the way. Then time the handoff to land when the eye is ready.

Place the product where the idea already lives

Once you have an attention route, you know where the product can actually sit. Write down three timings for it:

Recognizable. The moment a viewer could name it if asked.
Legible. The moment its role in the story is unmistakable.
Inhabited. The moment after the reveal, when it keeps doing a job.

If the product is only recognizable in the last second, you're showing a logo, not a relationship. If legible and inhabited are the same second, you have a cameo. If the route after the reveal has no idea left to carry, the back half becomes a tour of locations—nice to look at, doing no work.

Test the route by deleting the shot. If the product disappears and the sequence still makes sense as a story, the sequence is not about the product; it's a stunt with a brand taped to it.

Now the worked example: the parcel

This is an invented paper route, not a filmed sequence. It has not been shot, tested, approved, or costed. It exists so you can see what the choices look like on the page.

Premise for the passage. A small office. A sender at a packing bench seals a parcel and writes a mark on it by hand. The parcel moves through two intermediate pairs of hands—a bench hand and a courier—into the hands of a recipient who recognizes the mark. The relationship: the sender is the recipient's estranged sibling; the handwritten mark, a childhood shorthand, is the only personal thing in the parcel.

Attention route, written as an ordered list first:

Anchor: the sender's pen finishing the mark.
Handoff A: she places the parcel on the dispatch ledge; her gaze flicks to a bench hand at the far end.
The bench hand lifts it into a trolley.
Handoff B: the trolley's wheels, moving left out of frame toward reception.
Cut point to argue over: from trolley wheels to a reception counter, the courier signing the clipboard. In a single take, the move between these two spaces is the crucial staging decision; in a paper plan we can simply note that the two spaces must be clearly adjacent from the audience's point of view.
Handoff C: the courier takes the parcel and turns, parcel now in the foreground.
The corridor into the recipient's office. The parcel is briefly hidden behind the courier's body.
Handoff D: the courier sets it on the recipient's desk, mark-side up.
Recipient's hand pauses; thumb over the mark. The parcel is legible now; the mark is recognized; the recipient looks toward the doorway, and the shot keeps going as the courier's back disappears.

Two versions of the same route, to compare.

The route that reveals the relationship, written as a fragment of treatment prose:

She writes the mark. The parcel goes down the bench, and another hand picks it up. Through reception, the courier is in the way; we see his shoulder, not the box. He sets it on the desk with the writing facing up. The recipient's thumb stops over the ink. She doesn't open it. She looks toward the door where the courier went.

Every handoff above is a cue. The camera is not obliged to do anything clever; the staging makes the next attention point almost inevitable.

The same premise, the same route, and the same product, told as a camera tour instead of an attention plan:

The camera drifts from the packing bench, through reception and along the corridor, and settles on the recipient's desk, where the parcel is waiting.

The route is continuous. The viewer is continuously confused about what matters. The first fragment makes the parcel's movement and the mark's recognition a chain of knowing. The second makes the corridor the subject. That is the difference between a one-take idea and a camera move with a story dropped into it.

Where the cut point actually matters. In this example, the join between the dispatch trolley and the reception counter is the moment to argue about—the moment where attention has to be handled. If you cut it, you lose the sense that the parcel has been passed from one pair of hands to another as one act; the crossing between spaces becomes an editorial event rather than an event in the world. That is likely a loss in the relationship reading. But it is a reading, not a rule; in a different idea, a cut there would be the whole point.

True take or joined sequence—say which

Now the honesty part. There are two things people call a one-take:

True continuity. One unbroken recording. What the viewer sees is literally what the camera saw, start to finish. The continuity is real.

Continuous-looking construction. Separate recordings joined by a concealed edit—a wipe behind a passing body, a frame matched around a corner, a locked-off overlap. The continuity is an editing effect.

Neither is a failure. A hidden join is a tool. The problem is pitching one and delivering the other without saying so, because the proposal changes which dependencies you are asking collaborators to test.

Write it into the treatment, in language collaborators can act on:

The passage is intended to read as a single continuous move. The join between the dispatch area and reception is the focal point; the effect in front of the audience is a continuous move, while the filming method remains unresolved and will be tested with the camera and editorial teams.

Something like that is a production-aware proposal. It says what the viewer will experience and what the team still has to establish. It doesn't promise a true take or a disguised one before either has been worked out. It doesn't present a reference for one method as evidence for the other.

The distinction matters downstream. A true continuous take asks different things of staging, light, and performance stamina than a joined sequence does, and a proof for one proves nothing about the other. An ASC write-up of a noncommercial feature-length example can inform how physical endurance and available light behave over long unbroken running, but it can't stand as proof that a 45-second commercial can be achieved the same way. A feature-film account of apparent continuity shows that concealed joins can be invisible to an audience; it doesn't mean a commercial's schedule or talent conditions allow one. Keep those conversations separate, and say which the treatment assumes.

A rough route, plus the questions you can't answer alone

Before you pitch, produce two pages:

Page one: the action/attention route. Numbered handoffs, with the anchor, cue, and reveal noted for each. Mark the two or three handoffs whose timing or spatial relation is the least certain—the ones where, in your own head, the eye might miss the cue.

Page two: the questions. These belong to your collaborators, not to you, and should be written as questions rather than answers:

To the camera team: which handful of segments in this route need rehearsal to know whether the handoffs read from the intended angle?
To the design team: do the two spaces share a believable adjacency from the viewer's position, and where does that need to be demonstrated?
To the talent and props teams: can the parcel's mark stay legible from the intended positions so the recipient's recognition reads without a subtitle?
To the editorial team: which joins, if any, are we proposing to conceal, and what are we checking first?

None of those questions has a stunt answer, an equipment list, or a cost guarantee. That is deliberate. The treatment gives the team a legible creative dependency: what must be true for the idea to work, and where the unknowns are.

End where the route ends

Finish the treatment on the last handoff, not on the camera. State the route: the mark is written, the parcel goes through two pairs of hands, the thumb stops. Say what each handoff earns—the trust the recipient has to extend to accept the parcel at all—and stop. Do not end with a line about how beautiful it will be when the camera moves as one. That belongs to the idea's consequence, not its method.

If you want a test for the pitch before you write the closing line: ask whether the same document would still be coherent if somebody else's camera ended up doing it differently. If yes, you've pitched an idea. If no, you've pitched a camera move, and the audience is unlikely to have been the one it was written for.

Frequently asked questions

What makes a one-take premise worth keeping?

It should be able to name what the viewer loses if the sequence is cut. If the idea works just as well cut into six shots, the one-take premise has not earned its place. Three effects tend to be genuinely continuous: accumulation, an unbroken exchange, and a discovery. Continuity is not automatically immersive or authentic; it only produces the effect your staging and timing give it.

What is an attention route, and why is it different from describing camera movement?

An attention route is an ordered plan for handing the viewer's attention from one point to the next: anchor, handoff, and reveal. A description of where the camera drifts is a map, not a design; a moving camera does not guarantee anybody notices anything. The handoff is the hinge. A camera move that arrives without a cue leaves the viewer scanning the frame, and the intended continuity can feel like disorientation.

How can I tell whether the product is integrated into the idea or just taped to a stunt?

Track three timings: recognizable, legible, and inhabited. If the product is only recognizable in the last second, you are showing a logo, not a relationship. If legible and inhabited are the same second, you have a cameo. If the route after the reveal has no idea left to carry, the back half becomes a tour of locations. Deleting the product is a useful test: if the sequence still makes sense as a story, it is not about the product.

Should I pitch a true single take or a continuous-looking joined sequence?

Say which the treatment assumes. True continuity is one unbroken recording; continuous-looking construction uses separate recordings joined by a concealed edit. Neither is a failure, but pitching one and delivering the other without saying so changes which dependencies collaborators are asked to test. A true take asks different things of staging, light, and performance stamina than a joined sequence, and a proof for one proves nothing about the other.

What should a one-take treatment end on?

Finish on the last handoff, not on the camera. State the route and what each handoff earns, then stop. Do not end with a line about how beautiful it will be when the camera moves as one; that belongs to the idea's consequence, not its method. A useful test: would the same document still be coherent if somebody else's camera did it differently? If yes, you pitched an idea; if no, you pitched a camera move, and the audience is unlikely to have been the one it was written for.