Build a Cueable Playback List for a Small TV-Format Demonstration

A small TV-format demonstration usually comes down to four or five pieces of media and one or two moments where everything depends on a person. The playback list for that demonstration does not need a timeline. It needs named actions, and an honest answer for each one about what starts it.

There are only three answers. The operator presses GO. The next cue counts from the moment this cue began. Or the next cue waits until this cue finishes. Write one of those beside every cue and the list becomes readable in a dim room, during a missed press, by someone who did not build it.

The trigger definitions below come from QLab 5's public documentation on cue sequences. The arithmetic is paper reasoning about an explicitly declared zero-wait case — no workspace was built and no media was played in preparing this. That distinction matters more than it sounds, and I'll come back to it.

What follows is a local demonstration rig: one machine, one operator, one audience-facing output. It is not broadcast automation, it does not need a synchronized scoreboard, and it does not need a control network. If those things are on your list, put them somewhere else.

Name the five actions before you open the software

Start from the round you already agreed on. The most common way to break a small demo is to redesign the show while building the cue list, so that the media, the rules and the operator's sheet drift apart over an afternoon. Pick the round, then translate it.

The inherited example is a five-cue quiz round. It carries an introduction, an eight-second question clip, a neutral response hold while the participant thinks, the answer reveal, and a short closing sting. Here it is as an operator would read it.

That table is the whole design. Everything after this is detail about two of its cells.

Two constraints on the media. First, the hold card must not contain the answer in any form — not as text, not as a frame pulled from the reveal clip, not as a highlight sitting on the correct option, not as an audio bed that turns out to be the reveal sting. The unrevealed answer stays out of the public hold image, and the way to check is to look at the audience output during rehearsal rather than at the file on the desktop. Second, decide now whether the hold occupies the same surface as the question clip or a different one. If it shares the surface, Q3 replaces Q2 and the list is simple. If it does not, Q3's own action has to clear Q2 as well as bring up the hold card — that clearing is part of Q3, not a sixth cue. Nothing is still playing by then, because Q3 starts on Q2's completion; but a finished clip can leave its last frame sitting on the audience output, and forgetting to clear it is one of the small, stupid ways the question stays on the wall behind the hold card. Check the audience output in rehearsal to see whether your setup holds that frame.

Assign one operator. Print the sheet. Keep the trigger column visible on the printed page, because the operator's real job during a missed press is to find out what the software is waiting for, and that answer is in the trigger column.

Decide what starts the next cue

QLab gives a cue a trigger. It can be manual — you press GO — or automatic, and there are two automatic options with genuinely different meanings.

Auto-continue carries a wait that is counted from the moment the cue starts. Auto-follow starts the next cue after the current one completes. Those sound like two ways of saying "and then," and they are not. Manual is a third thing again and stays a third thing.

Here is why the difference bites. Take Q2, the eight-second question clip, and set it up with no waits at either end so its completion lands at eight seconds and nothing else is in play. Now compare two ways of handing off to Q3.

With a three-second auto-continue on Q2, Q3 begins three seconds after Q2 started. Q2 still has five seconds of question left. Whether Q3 stops the clip or sits on another layer, the transition happens while the question is running — the clip gets cut off, or two things are on screen at once. The three seconds is also a fixed guess. It matches the media only if the media is exactly three seconds long, and nothing in the cue list knows that.

With an auto-follow, Q3 begins when Q2 completes. The question finishes first. No guess, no overlap.

The trap is that a wait feels like control. It isn't. A three-second auto-continue does not cut a question short when a participant answers early — it cuts the question at three seconds, every single time, for everyone. If you want to cut a question short, that is a manual cue and a person deciding. Use the manual trigger for the judgment and reserve the automatic ones for things that are true of the media itself.

One more caution from the documentation's own housekeeping: some worked examples of timing arithmetic mix a wait with a total duration in a way that doesn't reconcile on inspection. Rather than copy numbers from an example that includes a wait you aren't using, state your own case plainly — zero waits, a known clip length — and verify the real timing against the real media later. Paper arithmetic tells you which trigger is which. It does not tell you how long your question clip actually is.

So in the five-cue list: Q1 and Q2 are manual. Q3 follows Q2 on completion, because the end of a prerecorded question clip is a real event that the software can detect. Q4 is manual, for reasons the next section covers. Q5 is manual too: the reveal stays up until the operator has seen the room read it, and a room finishing with an answer is not a media event.

Put the hold before the answer, not inside its timer

Here is the sentence this whole article exists to support: a prerecorded cue finishing does not mean a participant has finished.

That is why Q3 exists as a cue at all, and why Q3 does not end itself. While the participant thinks, the audience output shows the hold card and nothing else. No countdown, no music bed tightening, no subtle progress bar that tells the room the clock is nearly out. The hold ends when the operator decides it ends.

The operator's decision point is the answer-lock signal. That has to be a specific, agreed thing: the participant saying a fixed word, a host nodding at the operator, a physical card turned over. Whatever the format's rules say. Write it into the cue sheet using the same words the host uses out loud, because during the demonstration the operator is listening for those words and not reading a rulebook.

Keep two jobs separate here. Locking an answer is the game's business — whether an answer counted, whether the participant changed it, whether the host accepted it. Operating playback is the operator's business. The cue list is not the judge and cannot be. It has no idea whether an answer counted. It only knows whether Q4 has fired.

Then there is the handoff itself, which is where small demos actually fall apart. Q3 must end or be replaced before Q4 appears on the audience output. Decide how: either Q4 is authored to occupy the same surface Q3 was using, or Q4's own action stops or fades the hold card before the reveal appears. Either is fine. What is not fine is assuming it happened. Check it on the audience output — the screen the room is looking at — because the operator's workspace shows what the cue list is doing, not what the room can see. A hold card that reads as cleared in the list can still be sitting on the wall.

Two failure modes to name and avoid. Do not tie the reveal to an assumed response duration, so that a slow participant gets the answer thrown up before they have answered. And do not hide the answer merely beneath another window, minimized, behind the workspace, or on a second display the room can partly see. A reveal is either on the audience output or it is not; there is no safe middle.

Rehearse the path and write the reset

Preview plays a cue by itself. It does not walk the sequence. So if you have rehearsed by previewing Q2 and Q4 individually, you have not rehearsed the demonstration and you do not know whether the Q2-to-Q3 handoff works. Run the list on GO, in order, on the machine that will be used.

Three drills, each short:

The long response. Hold on Q3 for far longer than you expect — well past the point where the room gets restless. Does anything time out? Does the hold card survive? Does the operator have something to do, or only to wait? Waiting is a legitimate answer; discovering it during the demonstration is not.

The missed press. Have the operator deliberately do nothing. What is on the audience output? Is it safe to leave that way for thirty seconds? The operator should be able to work out from the sheet alone that Q4 is what they are waiting on.

The early press. Fire Q4 while the participant is still thinking. Watch what the room actually sees. This is the drill people skip and then need.

That third drill is where the reset procedure starts, because an exposed answer is a real outcome and it needs a real response. Reset is a separate operator procedure. It is not another timed cue and it should never be something the list does by itself.

Write it down. Stop playback. Clear the reveal material from the audience output. Clear anything pending or armed so an accidental press cannot bring back the reveal. Restore the entry state — Q1's slate up, the list selected back at the top, the operator's hands off the space bar. Then confirm on the audience output that entry content is showing, not black and not the reveal.

And one thing a reset cannot do. If the answer has already been on screen and the room saw it, stopping and restarting does not un-see it. Choose honestly: restart with fresh example content, or acknowledge out loud that the answer was shown. Rewinding is not concealment. Operators who try it lose the room faster than the mistake did.

What you have when you're done

Three artifacts, none of them a timeline: a cue sequence that runs from a named entry point, a written hold condition naming the exact answer-lock signal and the exact person who calls it, and a deliberate reset route that clears reveal material before entry is restored.

The list mixes manual and automatic triggers, and that mix is the point. Automation follows the media, because the end of a clip is knowable. The operator follows the people, because the end of a thought is not. Keep that line clear and the demonstration will hold up on the day, not just in the workspace.

Frequently asked questions

What three triggers can start a cue, and how do they differ?

Manual means the operator presses GO. Auto-continue carries a wait counted from the moment the cue starts. Auto-follow starts the next cue after the current one completes. Manual stays a third thing, not a variant of the automatic options.

Why is auto-follow usually right for moving from the question clip to the hold card?

The end of a prerecorded question clip is a real event the software can detect. Auto-follow begins the hold when that clip completes, so the question finishes first with no guess or overlap. A three-second auto-continue would begin three seconds after the clip started, cutting off the remaining question or putting two things on screen, and it would do that every time rather than responding to the participant.

What must the hold card not contain, and why?

The hold card must not contain the answer in any form: not as text, a frame pulled from the reveal clip, a highlight on the correct option, or an audio bed that turns out to be the reveal sting. The unrevealed answer stays out of the public hold image. Check the audience output during rehearsal rather than the file on the desktop.

What is the answer-lock signal, and what is the operator's role in it?

The answer-lock signal is a specific, agreed thing, such as the participant saying a fixed word, a host nodding at the operator, or a physical card turned over, according to the format's rules. Write it into the cue sheet using the same words the host uses out loud. Locking an answer is the game's business; operating playback is the operator's business, and the cue list only knows whether the reveal cue has fired.

What should a reset procedure do if the answer was exposed?

Stop playback, clear the reveal material from the audience output, clear anything pending or armed so an accidental press cannot bring back the reveal, restore the entry state, and confirm on the audience output that entry content is showing rather than black or the reveal. If the room already saw the answer, stopping and restarting does not un-see it. Choose honestly: restart with fresh example content, or acknowledge out loud that the answer was shown.

Further reading

• Build a Format-Demo Countdown That Handles Pauses and Late Updates — Read this if your format does intend a visible timer, since it chooses a time basis and freezes the display rather than repairing continuity from an unsupported number.