Turn an Unscripted Premise Into a Format That Can Run Again
Turn an Unscripted Premise Into a Format That Can Run Again
Suppose the idea on your desk is: two neighbors who have never really spoken build something together. Is that a show?
Not yet. It's a subject. It tells you where the camera points and roughly what the feeling should be. It does not tell you what happens on the day, or how you would shoot the same show again in March with two different households.
The distance between those two things is a procedure. A format names the conditions that stay identical every time — who is invited, what they are asked to do, what has to be finished before the session ends — and then leaves the people free to decide everything inside those conditions. Conditions repeat. Responses don't. What you are writing is the first half.
So the working answer to what makes this a format is: a stated set of conditions that recur, at least one decision the participants have to make inside them, a defined completion, and a clear account of what comes back next time and what doesn't. Get those four things on paper and you have something a production could attempt more than once. You still don't have something that works. That's a different document and a different question, and it gets asked later.
Find the activity inside the premise
"Two neighbors get to know each other" is an aspiration. Nobody can do it on cue, and it doesn't produce a decision a viewer can follow.
So ask a blunter question: what do these people physically do? Not what they feel. What they handle, choose, refuse, carry, build, negotiate or hand over. A making process with a fixed output does it. A negotiation over something there isn't enough of does it. A staged encounter with a deadline does it. So does a competition round — but nothing about the word format requires you to eliminate anybody. Elimination is one way to close a unit of television. It isn't the unit.
The useful test is whether something is insufficient. Not enough room, not enough time, not enough money, not enough information, not enough agreement. Shortage is what puts a decision in the room. A premise where everyone can have everything they want is a pleasant afternoon and a very hard edit.
This is where pitches go wrong in a recognizable way. A creator adds a prize, a scoreboard, or a weekly exit because the outline looks thin next to a show they've watched. Those additions create shape. They don't create activity. If the underlying task produces no decision, a scoreboard just records that nothing happened, in a more legible font.
Build one edition
Two things need specifying before you can write a second edition: what each person arrives with, and what counts as finished.
Here is an invented example, so you can watch the parts move. It is a paper format, not a tested one. Call it Shared Shelf.
Two households that live next door to each other on the same street are invited to fill one shelf in a street-facing window. Each household brings up to five objects from its own home. The shelf is a fixed size and holds six of the objects people tend to bring. The two households have one working session of a set length to agree on a single arrangement. For any object, either person may explain why they brought it, or say nothing. Nothing leaves the room while the session runs: anything that doesn't go on the shelf stays in frame beside it. The session ends when the shelf is arranged and photographed.
Notice how much of that is instruction and how little is hope. Each household brings up to five objects is an instruction. The pair must agree on one arrangement is an instruction. One of them opens up about an object is not an instruction, and never will be. You can build a rule that makes an explanation useful, and you should, because useful is different from required. But you cannot write a disclosure into a format and expect it at a convenient hour.
That distinction is worth holding onto, because it is where format documents quietly lie. A proposed turn can create the possibility that two people disagree in the room. It cannot promise the disagreement lands at minute thirty-two, and it cannot promise a reconciliation in the final beat. On paper, write what the rules make possible. Leave the outcome to the outcome.
Decide what resets and what carries forward
Run the invented format forward. The objects go home, all of them: the ones on the shelf and the ones beside it. The shelf is cleared. A different pair of neighbors arrives for the next edition. That's the reset, and it is what lets a viewer start anywhere.
What carries forward is the street, the window, and the photograph of the finished arrangement, which stays up in the corner of the window as the series continues. Nobody has to have watched an earlier edition to follow this one, though someone who has will notice the shelves have been accumulating.
Both choices cost something, and the costs are worth saying out loud.
Reset too completely and nothing accumulates. Every edition starts from zero, which means no edition is informed by the last, and the series is really a set of pilots sharing a title. Carry too much forward and a new viewer can't get in, at which point you have written a serial and should stop calling it a format.
The sharper test is what the second edition needs in order to be identical to the first. If the honest answer is "two households that haven't done this before," then the format is leaning on novelty rather than procedure. That survives once. A repeatable unit has to be a question you can ask again of different people, not a one-time event that happens to involve two of them.
Run two contrasting pairs through the same rules
Now the part that tells you whether you have a format. Put two very different participant situations through the procedure on paper, before you spend anything.
The first pair. Both households bring their full five objects and explain nearly all of them — where the thing came from, who it belonged to, why it stayed. That's ten objects and a shelf that holds six. Four objects have to come off. And because the pair has just explained why each one is there, the cut stops being a logistical question. It becomes an argument about which of ten stories gets to stand in a window on their own street. They negotiate. They use the whole session. They finish with an arrangement on the shelf and four objects on the floor beside it, in frame.
The second pair. One household brings two objects and explains neither. The other brings three and gives one-line answers. That's five objects against a six-object shelf. Nothing has to come off — there is no shortage. And because nobody explained anything, there is nothing to place against, no reason why the ceramic bird should sit left of the tin box rather than right. They put the objects up, agree quickly, and finish well before the session ends. The shelf is arranged and photographed, exactly as the rule requires.
That second pair is not a failure of personality. They cooperated. They completed the task. They are also an episode with no development in it, and the format is what allowed that.
Look at what the two runs exposed. The activity in the first pair came from two things: a shortage (ten objects, six places) and information (the explanations, which gave each placement weight). Both of those were left for the participants to supply. The rules permitted them. The rules did not require them. That is the gap, and it is a gap in the rules, not in the people.
Revise the conditions, not the people
The wrong fix is the tempting one: cast a pair who will talk more, or ask them to open up. That is not a format change. It's a note to a human being, and it makes recurrence depend on casting luck.
Move the rule instead. Two candidates, both with honest costs.
Require at least four objects per household, and require that at least one object from each household stays off the shelf. Now the set always exceeds the shelf — eight objects minimum against six places — and the cut always lands on both households, so both are genuinely in the decision. The cost: it forbids the outcome where one household cheerfully hands over all of its space to the neighbor. That might be the most interesting thing either pair does, and this rule makes it impossible. That's a real trade, not a technicality.
Add a revisit. After the display has stood for a set period, both households come back together and each may move one object. Now there is a second decision, and it arrives with the arrangement already standing in public view, which is different information than they had the first time. The cost: it roughly doubles the contact time you're asking for, and may move is still optional. A household that moves nothing has met the rule and skipped the decision, which is the first problem walking back in through a new door.
Neither revision is a solution. They are improvements with prices. That is roughly what a working format looks like: the rules put a decision where a camera can see it, and then the weight of that decision comes from somewhere the rules can't reach — who these people are, what they brought, and the fact that they will still be living next door to each other on Monday.
Write the mechanism at the size the pitch needs
A pitch does not need the working document. It needs two paragraphs: one describing a single edition as a viewer would watch it, and one stating what repeats.
The episode account says who is in it, what they are asked to do, what could change, how it ends, and what stays the same next time. Keep the unresolved choices visible. Write either may explain an object or not rather than pretending the pair will confide in each other. A reader who can see the option is a reader who trusts the rest.
Then be precise about which artifact you're handing over, because each one answers a different question:
- A paper walkthrough, like the two runs above. It tests whether the rules are internally consistent and whether a decision exists at all. It is invented, and should be labeled as invented.
- A proposed demonstration: a plan to run one session with one specific pair. It tests whether the procedure is executable in a room at all — timing, set, permissions. Still not evidence about people.
- A real test: an actual session with observed results. It tells you what happened once. Once is not a pattern, but it is the first real information you will have.
Operational documentation — consent, release paperwork, shot lists, timing sheets, the shelf's actual dimensions — belongs in its own document in its own phase. Loading an untested manual into the pitch is a way of looking further along than you are, and buyers tend to notice.
Why does the trade keep documents of this kind at all? Because a format has to be rebuildable by someone who isn't you. A record of FRAPA's services page, checked 8 September 2026, describes a Format Bible Generator among its resources, which is at least a sign that the industry treats a format as reproducible procedural knowledge rather than a subject line. Be careful about how much that gives you. It is a description of a resource, not the underlying template, and the record treats FRAPA and The Format People as the same initiative rather than two independent sources. It says nothing about whether any particular procedure works — including this one.
What the procedure looks like when you're done
The answer is a procedure, and it fits in a sentence. Shared Shelf, as revised: two neighboring households, minimum four objects each, one fixed shelf, one session, one agreed arrangement, at least one object from each household kept off the shelf, then a revisit where each household may move one object. That is the thing that repeats.
Run the two pairs again against the revised rules and watch what changes. The first pair still argues. The revision barely touches them. The second pair now has to bring four objects each and end the session with two of the eight beside the shelf, one from each household. Now the quiet pair is finally standing in front of a decision — though possibly one neither of them cares about, which is the honest limit of what a rule can do. Recurrence should show up as different human responses inside recognizable conditions, not as repeated promises that something dramatic will happen.
Then write down what a test has to answer, because that list is what you'll work from next:
- Do real participants bring the number of objects the rule assumes, and does the combined set reliably exceed the shelf?
- When there is nothing to cut, does anything else in the procedure produce a decision?
- Does the optional explanation get used — and when it doesn't, does the arrangement change, or only the conversation?
- How long does a session actually run, and what is on camera when it runs short?
- Does either household's relationship change once the arrangement is standing in public, and does that belong in this format or in a later one?
That's a procedure, two contrasting runs, and an unanswered list. It isn't a show yet. It's the document that lets you find out whether it could be one.
Frequently asked questions
What turns an unscripted premise into a format?
A format needs a stated set of conditions that recur, at least one decision participants have to make inside them, a defined completion, and a clear account of what comes back next time and what does not. That is what a production could attempt more than once; whether it works is a later question.
Why does shortage matter in a format?
Shortage—not enough room, time, money, information or agreement—is what puts a decision in the room. A premise where everyone can have everything they want is a pleasant afternoon and a hard edit. Adding a prize, scoreboard or weekly exit can create shape, but it does not create activity if the underlying task produces no decision.
What does Shared Shelf show about reset and carry-forward?
In the invented example, objects go home, the shelf is cleared and a different pair arrives next edition; the street, window and photograph carry forward. Reset too completely and nothing accumulates, making editions feel like pilots sharing a title. Carry too much and a new viewer cannot get in, which is a serial rather than a format. If the second edition needs two households that have not done this before, the format is leaning on novelty rather than procedure.
How should a creator fix a pair who completed the task with no development?
Revise the conditions, not the people. Two candidates are requiring at least four objects per household and requiring at least one object from each household to stay off the shelf, which costs the outcome where one household gives all its space to the neighbor; or adding a revisit where each household may move one object, which roughly doubles contact time and still allows a household to skip the decision.
What pitch documents does the article distinguish?
A paper walkthrough tests whether rules are internally consistent and whether a decision exists; it is invented and should be labeled as invented. A proposed demonstration plans one session with one specific pair and tests executability—timing, set, permissions—but is not evidence about people. A real test is an actual observed session; once is not a pattern but is first real information. Operational documentation belongs in its own document and phase. A pitch needs two paragraphs: one describing a single edition as a viewer would watch it, one stating what repeats.