Share a Commercial Pitch With Specialists Without Sending Everyone the Whole Brief
Share a Commercial Pitch With Specialists Without Sending Everyone the Whole Brief
Give a specialist the smallest authorized package that still contains the conditions their answer depends on. That is the entire job, and most of the difficulty lives in the word "conditions."
There are two ways to get this wrong, and they fail in opposite directions. Send the whole client package to everyone who touches the pitch and you have disclosed material nobody asked for and possibly nobody approved — and, in practice, the specialist skims it and answers whichever part happened to catch their eye. Send a tight, well-written excerpt that quietly drops one constraint, and you get something worse than a slow answer: a confident, careful, well-reasoned recommendation that was never actually available to you.
The second failure is the one that hides. Nobody notices it until the shoot.
Here is the test worth keeping. An omission is material when it makes the specialist treat a fixed variable as free. Not "would they find it interesting." Not "would it be nice for them to know." Would their answer change — or would they design around a freedom that does not exist?
Write the question before selecting the excerpt
An excerpt is the answer to "what do they need," and that question has no answer until you have written down what you are asking. Two sentences is usually plenty:
- What the specialist is being asked to assess.
- What sits outside their contribution.
A request to judge whether a hand-to-product action reads on camera needs different context from a request to explore an animated transition, even when both concern the same four seconds of the same spot. Whoever is judging the move needs physical facts: what the performer does, what the product behaves like, how long the move takes, what the frame looks like at the end. Whoever is designing the transition needs the boundary: which parts of the plate are real, which are proposed, and whether the result has to survive at phone size.
Three things belong in the question itself. The proposed action, stated plainly. What cannot change, stated as a constraint rather than a preference. And the shape of the answer that would help — a yes or no with a reason, a redesign, a conditional, a rough cost.
The temptation to skip this is real, because forwarding the project folder is easier than defining a question. It also moves your work onto the specialist's desk. Handed the whole package, a specialist will often answer the loudest part of it, because the loudest part has the most material attached to it, not because it is the part you need solved.
Say what is out of scope too. "We are not asking whether the client will approve this" saves a paragraph of hedging on both sides.
Keep the conditions that could change the answer
Material conditions hide in a handful of predictable places:
- Scale and playback. A label that reads on a monitor and a label that reads on a phone are different briefs.
- Timing. Durations, cut points, how long a thing is on screen at its smallest.
- The physical product. What it is made of, how it behaves under a hand, what is loose inside it.
- Performance. What a person may and may not do on camera during the shot.
- Status. Whether the thing under discussion is approved, proposed, or one of two routes still competing.
That last category gets dropped more than any other, and it is the one that most reliably changes the answer. A specialist who thinks a proposal is locked will design against it instead of questioning it.
An excerpt that keeps only the interesting middle of an action hides the entry and the exit, and entry and exit are frequently what decide whether the middle works. How something starts and how it finishes are not decoration around the question. They are often the question.
Then say what you left out, and say where it matters. There are two honest versions of this line:
Not included, because it does not change this answer: the music strategy, the back half of the spot, the retail timeline.
Not included, because we do not have permission yet: [the thing]. Answers to this question will be limited accordingly.
The second version is the one people skip, and skipping it is how you get an unqualified recommendation built on a view you were not entitled to give. You can name withheld material at the level of category — "internal review notes were withheld" — without exposing contents.
Check the recipient, use, and transfer route
Three separate checks, and they do not collapse into each other.
Who. A named person. Being obviously the right person technically is not the same as being on the list the owner approved for this project. Need-to-know is an argument for asking; it is not the permission itself.
What for. This question, this task. Not "the pitch." A broad grant covers uses you did not intend, including ones that arrive later — a reel, a portfolio, a second opinion from a friend who happens to be good at this.
How. The route. A link with access limits and an expiry is not the same as an attachment in a group thread, and the owner may have a settled view about which is acceptable.
Two things stay separate throughout. Internal approval and technical judgment — someone signing off on the share does not make the specialist's geometry correct, and a specialist being obviously correct does not make the share authorized. And permission to read a constraint is not permission to read the file it came from. If the excerpt says the label type follows the packaging dieline, that is a reference. It is not an instruction to attach the dieline. Send the constraint, or ask.
Whether client material may enter a third-party tool at all is a different check from the one in this article, and a stricter one. This piece is about the human on the other end.
Escalate only the missing context
When the excerpt genuinely cannot carry a dependency, ask for that specific thing — and give the reason in terms of the answer you need. "The animation lead cannot estimate the tracking without knowing the pouch film flexes during the move. Can we share that one line from the brief?" A specific request with a stated reason is far easier to approve than "can we send more."
If permission does not come, you have a ladder, and it is worth using it in order rather than reaching for the top rung:
- Narrow the question so the authorized material supports it. "Set aside the flexing film — can your method hold a planar track on a rigid face at all?"
- Take a conditional answer, with the assumption written down.
- Leave it unresolved, and say clearly that it is unresolved.
A conditional answer earns its place if it carries three things: the assumption, what would change the answer, and who owns closing the gap. Without those, a conditional answer is just a guess wearing a lab coat.
The two failure modes here are symmetrical. Indiscriminate disclosure on one side; on the other, redaction so aggressive that no responsible answer remains on offer. The second is more dangerous, because it looks like diligence and reads as progress.
One brief, three excerpts
Everything below is fictional. It exists to show the shape of the problem, not to report a real job, a real permission, or a real result.
The invented brief. A 20-second spot for a dried-citrus pouch brand. The hero section runs four seconds. A performer seated at a small table lifts the pouch and carries it toward the lens in one continuous move lasting about 2.5 seconds, ending with the pouch centered and filling roughly a third of the frame height against a dark matte wall. The label is a printed card bonded inside the pouch's front film face, carrying the brand wordmark and one line of copy; the wordmark must stay legible at phone playback. At the lift, the pouch face sits about 35 degrees off the lens axis, and nothing in the move turns it. The key light is a large soft source camera-left, roughly 45 degrees above the table — not yet locked. Two routes are still under consideration: label physically in the pouch at the shoot, or a blank pouch on the day with the label added in post.
Three specialists, three questions, three excerpts.
The cinematography excerpt, and the line that decides it
Here is the version of the excerpt that supports a valid answer.
Question: Can the lift and approach work as one continuous 2.5-second
move, ending with the pouch at a third of frame height and the wordmark
legible at phone size — and if not, what would you change?
Fixed:
- One take, no cut.
- The hold established at the lift is the hold at the end. No wrist
roll, no re-clasp during the move. This is a treatment decision:
any adjustment of the grip mid-move reads as fidgeting.
- Final frame: pouch centered, roughly one third of frame height.
- Wordmark legible at phone playback.
In this excerpt: the move, the pouch and label construction, the
final framing, the dark background, the general key light position,
and the pouch face angle at the lift.
Not included, because it does not change this answer: the music
strategy, the back half of the spot, the client's retail timeline.
Assumptions: key light position is not locked. Answers that depend
on it should say so.
What would change our answer: a locked light position, or a change
to the final framing.
Now delete the "Fixed" line about grip and add nothing in its place. That is the deliberately incomplete version. It still reads cleanly. It still looks like a professional brief. The only signal that something is missing is that there is no signal at all.
Ask a cinematographer the incomplete question and a reasonable answer comes back: yes, the move is ordinary — push in along the approach, roll the pouch about 30 degrees in the last half-second to square the face to the lens, and the wordmark lands clean. That answer is not wrong about the geometry. It is wrong about the job, because it spends a freedom the treatment has already taken away.
Ask the complete question and the answer changes shape. The face enters the final frame still off-axis, and nothing in the move corrects it. The fixes now live outside the move: change the hold before the lift, which the constraint permits because it governs the move and not the setup; or move the camera around a fixed hand rather than pushing straight in; or plan the label as a post element and shoot the pouch clean. The specialist may also come back with a way to turn the face that is not a wrist roll — a shoulder or forearm adjustment that satisfies the treatment's actual intent. That is a better answer than the first one, and it only becomes available when the constraint is visible.
One line. That is the size of the difference between a useful recommendation and a confident dead end.
The other two excerpts
Animation, proposed transition. The question here is whether the label can be tracked onto the moving pouch face in post and then used as a wipe into the end card. The excerpt carries the move's final 1.5 seconds, the pouch and label construction, and the requirement that the label read as printed paper bonded to the inside of the film. It also carries a status line the first draft had left out: the physical-label and post-label routes are both still live, and the animation team is evaluating the second, not the first. Without that line, an estimated tracking cost quietly becomes a commitment to a route nobody has chosen.
The dependency it does include is a physical fact: the pouch film flexes as the hand moves and the loose contents settle, so the face the label tracks to is deforming. That is not a permission item. It is a line from the brief that got dropped as "production detail," and without it the animation answer assumes a rigidity that does not exist.
Research, observed transparent surfaces. This excerpt asks whether the observations the researcher already has support a general statement about contrast loss over a printed card behind a clear film at oblique angles. The material and the approximate geometry are in the excerpt. The key light position and the final camera roll are not, because they are not locked. So the honest answer comes back conditional: at the lighting tested, contrast falls when the source's reflection crosses the label; whether it crosses in this shot depends on a position nobody has fixed yet. The excerpt says so, and says who owns closing it.
What this comes to
The three excerpts share a shape, and it is not a form to fill in so much as a habit of mind. Each one states its own question. Each one states what is fixed. Each one states what it left out and whether that omission limits the answer. Each one names the assumptions the answer will rest on. And each one records who authorized the share, for what task, over which route.
The result is not the smallest possible disclosure. It is the smallest disclosure that still supports a true answer, which is a different and slightly larger thing. A brief that removes unrelated client information is doing its job. A brief that removes a condition which changes feasibility is producing a fiction with a professional finish.
So end where the work actually is. Send the focused excerpt, with its assumptions written down where the specialist can see them. Keep the withheld list honest, including the part that says what you could not get permission to include. And if the excerpt still cannot carry a necessary dependency, do not paper over it — go back to whoever owns the material and ask for that one thing, with the reason attached. If the answer is no, narrow the question to what the authorized material can support, or leave it open on purpose.
A smaller package is a success only when the question inside it remains both truthful and answerable.
Frequently asked questions
What makes an omitted condition material rather than just interesting?
It is material when its absence would make the specialist treat a fixed variable as free — when their answer would change, or they would design around a freedom that does not exist. Interest or niceness is not the test.
Why write the question before choosing the excerpt?
An excerpt is the answer to what the specialist needs, and that cannot be determined until you state what you are asking. Two sentences usually cover what they are assessing and what sits outside their contribution; the question should include the proposed action, the constraint stated as a constraint, and the shape of answer that would help.
Which conditions are most likely to change a specialist's answer?
Scale and playback, timing, the physical product, on-camera performance, and status — approved, proposed, or competing. Status is dropped most often and reliably changes the answer; entry and exit of an action can decide whether the middle works.
How should withheld material be handled?
State what was left out and whether it limits the answer. Two honest versions are: not included because it does not change this answer, or not included because permission has not been granted, naming the category if contents cannot be exposed. Skipping the second version produces an unqualified recommendation.
What should you do if permission for a needed dependency is refused?
Use the ladder in order: narrow the question so authorized material supports it; take a conditional answer with the assumption, what would change the answer, and who owns closing the gap; or leave it unresolved and say so. A conditional answer without those elements is a guess.