Product

What the composer decides for you, and how to override it

Format Short dramaMarket Global English
Short answer

Five layers — product, actor, format, hook and setting. Anything you leave unset resolves through Director's Choice, where the engine fills it against your prompt rather than omitting it. So nothing is skipped; it's decided for you. Only three things affect credits: length, resolution, and whether you select an actor. Prompt detail and layer choices are free.

The useful thing to understand about a composer isn't what it generates. It's that leaving a field blank is a decision rather than an absence.

What are the five layers?

A scene prompt, plus five modular layers arranged around it.

Product — what's being shown or sold. Actor — who's on screen, or nobody. Format — the shape the piece takes. Hook — how it opens. Setting — where it happens.

Each is configurable individually, and each can be left alone. The prompt is where the specificity goes; the layers are where structure gets set.

Leaving a layer blank doesn't remove it. It hands the decision to the engine.

That's Director's Choice, and it's the behaviour worth knowing before your first generation. An unset layer resolves against your prompt — so the engine reads what you wrote and fills the gap with something consistent with it, rather than producing a piece with no setting or no hook.

What does it decide, and how do you take it back?

Layer by layer, with what happens when you don't set it.

LayerLeft unsetHow to take it back
ProductInferred from your prompt. If the prompt doesn't name one, you get whatever the prompt impliesName it explicitly, and describe it rather than referring to it
ActorChosen for you, or resolved as faceless depending on the promptSelect one from the library, or choose faceless deliberately. This is the only layer that changes the price
FormatResolved against the prompt's apparent intentSet it. This is the layer most worth setting, because it determines the piece's whole shape
HookThe engine writes an openingWrite the opening in the prompt. A hook you wrote beats a hook that was inferred
SettingFilled to match the promptDescribe the location, or attach a saved location reference

Source: Hexcoded Marketing Studio documentation and pricing page, checked September 2026. Tool behaviour changes — verify against the product.

Where this falls short. Director's Choice resolving well depends entirely on the prompt being specific. A vague prompt gives the engine little to resolve against, and the result is generic in a way that reads as the tool's fault rather than the brief's.

What actually costs you credits?

Three things, and it isn't what most people assume.

Length, which is proportional. Resolution, which rises by tier. And selecting an actor — casting a real human creator adds a small flat fee, while AI actors and an actor you made from your own photo don't.

What costs nothing: prompt length, the format, hook and setting selections, custom layer phrases, and inline asset tagging.

Creative specificity is free. Scale isn't. That's an unusual shape and it should change how you work.

Read that properly and it inverts the normal instinct. There's no reason to write a short prompt to save money, and no reason to leave layers unset to keep costs down. Detail is free.

What costs money is generating at length and at high resolution — which means the efficient workflow is maximum specificity at minimum scale while the creative is still moving, then scale up once it's locked.

The cost also shows before you render, so none of this has to be estimated.

What are the other settings?

Two, and both are set rather than inferred.

Length runs four to thirty seconds on a slider, defaulting to fifteen. Unlike Talking Actors — where duration is derived from the script and can't be set — here it's a direct control.

Aspect ratio offers six options, so this isn't a vertical-only tool despite vertical being the common case.

Plan First generates a storyboard and script preview without consuming credits. For a tool that fills in your blanks, that's the place to check what it filled them with before spending anything — which makes it more valuable here than in tools where you set everything yourself.

When should you let it decide?

Two cases, and they're both legitimate rather than lazy.

When you don't have a view. If the setting genuinely doesn't matter to the piece, specifying it is work with no return. Director's Choice resolving it against a good prompt will produce something coherent.

When you're exploring. Early passes benefit from the engine making choices you wouldn't, because you find out what the prompt actually implies. That's information you can't get by specifying everything.

The case against is narrower than it sounds: leave layers unset on a piece that has to match something — a series look, a brand guideline, an existing episode — and you're gambling on the engine inferring a constraint you never stated.

When should you set everything?

When consistency matters more than exploration.

A series is the clear case. Episode 40 has to match episode 3, and Director's Choice resolves against your prompt rather than against your previous episodes — so consistency has to come from what you specify rather than from what the engine remembers.

The mechanism for that is the same one Creative Studio uses. Save your actor, your setting and your look as named references, then call the handle rather than re-describing them. A saved reference persists; a re-description varies.

1

Write the prompt as specifically as you can

Prompt length is free, so there's no reason to compress. Everything the engine has to infer is a decision you didn't make.

2

Set format first, then actor

Format determines the shape of the whole piece, and actor is the only layer that changes the price. Those two are worth deciding deliberately even when you leave the rest.

3

Run Plan First

Free, and it shows you what the engine filled in. On a tool that resolves your blanks, that's the cheapest possible place to catch a wrong inference.

4

Iterate short and low, then scale

Length and resolution are the only drivers that scale with cost. Work out the creative at fifteen seconds and a lower tier, then generate the keeper at what you need.

The bottom line
  • Five layers: product, actor, format, hook, setting. Anything unset gets filled rather than skipped
  • Director's Choice resolves against your prompt, so a vague prompt produces generic fills
  • Only three things cost credits: length, resolution, and selecting an actor
  • A real human creator adds a small flat fee. AI actors and one made from your own photo don't
  • Prompt length, layer selections and inline tagging are all free. There's no reason to compress a brief
  • Iterate short and low, then scale. That's the only workflow the cost structure rewards
  • Length is a slider here, four to thirty seconds, unlike Talking Actors where the script decides
  • For a series, set everything and use saved references. The engine resolves against your prompt, not your last episode

Any of the five layers you leave unset — product, actor, format, hook or setting. They resolve through Director's Choice, where the engine fills them against your prompt rather than omitting them. So a blank field is a decision handed over rather than an absence.

The behaviour that resolves unset layers against your prompt. It reads what you wrote and fills the gap with something consistent with it, which means the quality of the fill depends entirely on how specific the prompt was.

Three things. Length, which is proportional. Resolution, which rises by tier. And whether you select an actor — a real human creator adds a small flat fee, while AI actors and an actor made from your own photo don't. Prompt length, layer selections and inline tagging are free.

No. Prompt length, custom layer phrases and inline asset tagging cost nothing. Creative specificity is free and scale isn't, so there's no reason to compress a brief to save credits.

Four to thirty seconds on a slider, defaulting to fifteen, across six aspect ratios. That's different from Talking Actors, where duration is derived from the script and can't be set directly.

For a series, yes — the engine resolves against your prompt rather than against your previous episodes, so consistency has to come from what you specify. For exploration, leaving layers unset is useful, because the engine makes choices you wouldn't and that tells you what your prompt actually implies.

Describe it, or decide it

Five layers you can set or leave. Prompt detail costs nothing, only length and resolution scale, and Plan First shows you the storyboard before it spends a credit.

Make a video

More on tools, features and how the platform fits together in Product.