What Can the Audience Actually Change in Your TV Format?

By pitch.dogFoundation12 min read20 September 2026

Television

A pitch that calls a show interactive is making a claim with a shape: somebody does something, and something in the program is different because of it. If you can't finish that sentence — who acts, and what changes — you don't have a mechanism yet. You have an atmosphere. "Engagement" is a mood word. It sounds like an answer without committing to one.

So the first job isn't picking a platform or naming a feature. It's naming the action.

Name the audience's action and the authority it carries

Write the sentence as plainly as you can: [viewer action] changes [part of the program]. Then check which of four things you've actually written, because they carry authority of different sizes, and mixing them up is where pitches start lying without meaning to.

Playing along. A viewer guesses the answer, the show reveals it, the viewer compares. Nothing in the program changes because of the guess — not the outcome, not the edit, not the guest's next line. The authority is real, but it's over the viewer's own evening. That's a legitimate pleasure and a cheap one to build, but it isn't participation in the program, and a pitch that presents it as participation has borrowed authority it can't spend.

Contributing material. Viewers send something — a story, a photograph, an object, a recording. The program is different because that material exists; without it, there is no segment. But the audience doesn't choose what appears. Production sorts the pile, edits it, decides which of forty submissions fills ninety seconds. Authority over supply, not over selection. "Some of what you send may be on the show" is an honest sentence. "You decide what we build" is not.

Supplying material and choosing among supplied material are two different acts, and both get filed under the word "submissions." If the audience picks which of the forty gets made, that's the third rung in shape — a selection among options — with one structural difference: the pool isn't production's. It arrives from viewers, and production can only predeclare what it can still fix before the window opens: eligibility, removal categories, the threshold and tie rules, and who rules on disputes. Declared in advance, those rules bind. But the option set is no longer the producers' to commit, and a pitch that claims it is has borrowed authority from a rung it isn't standing on. Who fixes the pool is the whole difference between two categories that both get called submissions.

Choosing among predeclared options. Production commits the option set in advance — two materials, three endings, four routes through the build — and every option is genuinely feasible, stocked and shootable. The audience's action selects one. Production keeps the fence; the audience chooses inside it.

Determining a result. Production commits, before the tally is known, to execute whatever survives the rules — including an outcome the producers would not have picked. That commitment is what separates deciding from choosing, and it has nothing to do with how many people took part. Four thousand viewers choosing between two things the show would happily do either way have chosen. They have not decided. If there's a veto the producers would use, the audience chose and the producer decided.

Notice that the jump between the third and fourth rungs isn't a matter of enthusiasm or volume. It's a promise made in advance, when the result is still unknown and might be inconvenient.

Whatever rung you're on, some decisions stay with production: the option set, who counts as eligible, how ambiguous input gets read, what happens when the rules don't cover the case, and when the whole thing happens. On the open-pool version of contributing material, the option set is the item that isn't yours — viewers supplied it — so eligibility, the removal categories and the tally rules have to carry the predeclared authority instead. Everything else on that list you can keep. What you can't do is keep them quietly and still use the word "control." If the show can override the outcome, the on-screen language should not say the audience decides it.

And resist the temptation to call all four "interactive." The word flattens exactly the distinctions a commissioner needs.

Locate the input in the episode's actual timing

Here is a worked example, invented for the purpose. It's a design sketch, not a production: no participation service, voting system or audience test stands behind it. Call the show Made Here — thirty minutes, weekly, Monday at 20:00, a guest maker building one object. Episode 3 ends with an offer. Two materials are predeclared on air: cast concrete and turned oak. Both are already in the workshop, both are tooled and safe to shoot in the same slot, and the guest can work either.

The timing, stipulated:

The request appears at 20:26 in Episode 3's closing segment. The host names both materials and the route for choosing.
The window closes Thursday at 23:59 — roughly three days later.
Processing runs Friday, 09:00 to 15:00. A designated producer reviews submissions, applies the predeclared removal rules, tallies what remains, and rules on anything ambiguous or disputed.
The choice locks Friday at 15:00 and is written down. The guest is told that afternoon and rehearses over the weekend.
The effect becomes visible Monday at 20:10 in Episode 4, when the guest picks up the material and starts the demonstration.

Those four marks are the minimum a pitch should carry, and the gap between "closes" and "becomes visible" is not dead time. That gap is where moderation, tallying, tie rules, guest prep and rehearsal all live. Shorten it and you don't get a leaner format; you get fewer chances to notice that something has gone wrong.

This example is deliberately delayed: the choice applies to a later demonstration, collected after the broadcast. A live version is a different mechanism, not the same one with a tighter clock. Live means the window has to close, the tally has to resolve, and the result has to be safe to broadcast inside a single transmission — and it means you need a fallback that works at the moment airtime arrives, whether or not the numbers have resolved. Intended live action and later contribution fail in different ways. Pick one on purpose.

Two cautions about the numbers above. They are stipulated, not researched. Broadcaster participation terms vary by show, market and edition, and the fallback, cutoff and replay rules for the format you're actually pitching may not resemble these at all. Read the terms for the named edition before you write a single implementation sentence.

Explain how usable inputs become a program decision

The audience's action doesn't become a program decision by itself. Something has to convert it, and the pitch should say what.

The rule. For Made Here: simple majority of usable submissions between two materials. Everything is written down before the window opens, including the awkward cases.

The threshold. A predeclared minimum below which no choice is made. Say the pitch sets it at 500 usable submissions. Whether 500 is the right number is a scheduling and confidence question for the team, not a fact about the audience — the number only matters because it decides whether a thin response can bind the episode.

The tie. With two options, an exact split is possible. Predeclare it. In this design, the default wins ties, which is slightly conservative and easy to explain on air.

Moderation. Removal categories: submissions outside the window, bulk or automated entries, write-ins that name a third material the show can't build, and anything that can't go on television. One important clause — removed submissions count toward neither the tally nor the threshold. Without that clause, a flood of junk can carry a result over the line, or a thin night can bind the episode on nine usable votes.

Duplicates and second thoughts. The design has to choose: does the last submission from a person count, or the first? "Last counts" lets someone change their mind up to the close, which is friendlier. "First counts" is simpler to audit. Either is fine; both change what the final number means, and neither is fine if it goes unstated.

Who resolves disputes. A named role, off camera, whose ruling is written before the window opens — so the rule can't be reverse-engineered from a result that arrived inconveniently. If the host rules on camera, you've built a competition with an unannounced adjudicator.

Feasibility of both options. Predeclared means prepared. If the oak supplier falls through midweek, don't quietly substitute walnut; a substitution nobody announced turns a real choice into a staged one. Either a predeclared alternate takes oak's place and the audience is told, or the default applies.

Keep all of these as design decisions until the people responsible for the build confirm them. And be careful with what a majority proves. A majority of submissions is a majority of people who submitted, not a majority of viewers, not a sample of anyone, and not evidence of what an audience will enjoy watching. Popularity doesn't establish representativeness, fairness or suitability for broadcast. Sometimes a result the audience chose still can't air as chosen — a safety problem, a rights problem, an accessibility problem. If a fallback exists for that, it should have been predeclared too, and the episode should say it happened.

Design the closed-window and no-input versions

Three things go wrong, and they are not the same thing.

Nothing arrives. Too little arrives — under the threshold. The mechanism fails — the submission route is down for a day, or a processing step doesn't work. The first two are outcomes; the third is a fault. Extending a window because of a fault can be the right call, but only if the extension rule was predeclared. Deciding to extend after you've seen the shortfall isn't a rule, it's a preference.

For all three, the example needs a default named in the same breath as the options: if no usable choice is made, the guest works in concrete. The episode still stands. The demonstration still happens. Nobody has to fill airtime explaining why the thing the show promised didn't occur.

Now the part that's easy to get wrong. The credit line at the 20:10 reveal has to distinguish itself from the default. "Chosen by viewers in the week of [dates]" and "No qualifying choice received; this is the predeclared default" are different sentences, and if concrete wins with 4,000 submissions and also happens to be the default, the audience still deserves to know which one happened. Otherwise the default quietly launders itself into an audience decision, and the format has borrowed authority it never received.

Later viewers need the same care, and they get three honest treatments.

Show the recorded result, labelled closed. The on-demand viewer sees what the audience chose and is told the window it came from has shut.

Offer private play-along. The viewer picks before the reveal, then sees how the original audience went. That's a guessing game, and it's honest as long as it's presented as one.

Omit the invitation. For the on-demand cut, drop the closing-segment offer or replace it with a note that the window is closed. What you must not do is leave "you decide" in a two-year-old episode for a decision that closed two years ago.

Those are three different programs. A replay isn't automatically interactive just because the original transmission was.

Before any of this reaches a commissioner: this article's example is invented, and no vote, moderation system, eligibility rule or audience test exists behind it. Two pieces of work belong in front of the pitch. Read the participation terms for the broadcaster, show and edition you're targeting, and find one episode account that shows the input's effect, its cutoff and its replay treatment — separately, from the named edition, not from a general impression of how platform features usually work. Then get technical, accessibility and legal review for any live mechanism. A conceptual design isn't operational approval, and a pitch never fails more expensively than when it has already promised the audience control in a sentence the production team can't keep.

Which brings it back to one description, the one that should be legible to a viewer who has never read the pitch: viewers choose which of two predeclared materials a guest uses in a later demonstration. The invitation airs at 20:26 in Episode 3, the window closes Thursday at 23:59, submissions are reviewed and tallied Friday between 09:00 and 15:00, the choice locks at 15:00 the same day, and the guest learns it that afternoon. Episode 4 shows the material at 20:10 on Monday, labelled with the week it came from — or, if the threshold wasn't met, labelled as the predeclared default. A viewer arriving later sees the recorded choice and is told the window has closed, or gets a private guess before the reveal that never pretends to decide anything. That's a promise a format can keep, and a viewer can trace.

Frequently asked questions

What sentence must a pitch be able to finish to show an interactive mechanism?

It must be able to say: [viewer action] changes [part of the program]. That names who acts and what changes. Without it, 'engagement' is only an atmosphere rather than a committed mechanism.

What are the four levels of audience authority, and how do choosing and deciding differ?

Playing along changes only the viewer's own evening. Contributing material supplies content, but production selects what appears. Choosing among predeclared options lets the audience select one of a production-committed, feasible set inside a fence. Determining a result means production commits before the tally to execute whatever survives the rules, even if producers would not have picked it. If producers would use a veto, the audience chose and the producer decided.

In the delayed example, what timing marks should a pitch carry, and why does the gap matter?

The request appears at 20:26 in Episode 3; the window closes Thursday at 23:59; processing runs Friday 09:00 to 15:00; the choice locks Friday at 15:00; and the effect appears Monday at 20:10 in Episode 4. The gap between close and visible is where moderation, tallying, tie rules, guest prep and rehearsal live. Shortening it reduces chances to notice something has gone wrong.

What rules have to be predeclared so usable inputs become a program decision?

They include the rule, threshold, tie handling, moderation and removal categories, treatment of duplicates and second thoughts, who resolves disputes, and feasibility of both options. Removed submissions should count toward neither the tally nor the threshold. Disputes should be resolved by a named off-camera role whose ruling is written before the window opens. If a predeclared alternate is needed, the audience should be told.

How should the default and later viewing be handled?

A default should be named with the options, such as the guest works in concrete if no usable choice is made. The credit line must distinguish an audience-chosen result from the default even if the same outcome occurs. Later viewers can see the recorded result labelled closed, get a private play-along that is presented as a guessing game, or get no invitation or a note that the window is closed; they must not be told they decide a decision that already closed. The example is invented; no vote, moderation system or audience test stands behind it, and the terms for the named edition still need to be read.