How Much of Your Visual Novel Should You Show First?

You have a finished first chapter, a store page that needs something playable, and a question that sounds as if it should have a numerical answer: how much of your visual novel should you show first?

Search for an answer and you will find every number. Twenty minutes. An hour. Ten percent of the full game. The whole common route. Each answer can describe an opening that worked, but none tells you whether it fits your story. A mystery can establish its central question in twelve minutes and still need another half hour before the reader makes a meaningful deduction. A romance can introduce everyone quickly and remain emotionally inert until two characters are finally alone. The clock records what happened; it does not tell you when the promise of the game became clear.

Valve's own Steam demo guidance avoids giving developers a target duration. It describes a demo as a small playable portion that shows the core experience, then asks creators to balance enough content to create interest against so much content that the player feels finished. Steam Next Fest goes further: it requires a public playable demo for participating games, but explicitly says there is no length requirement. That is useful permission. Your stopping point is a design decision, not compliance with a hidden platform rule.

The practical answer is to stop measuring the first playable section and start identifying what it has to prove.

What must the opening prove?

The opening of a visual novel makes three promises at once.

The story promise answers: what kind of tension will keep moving this forward? It might be a disappearance nobody is willing to discuss, a relationship that becomes impossible to ignore, or a job that asks the protagonist to compromise something important. This is not the entire plot. It is the pressure that will keep producing scenes.

The character promise answers: who will the reader want to spend time with? A cast portrait and five introductions do not necessarily satisfy it. The reader needs at least one exchange where a character's public role slips and a more specific person appears underneath.

The interaction promise answers: what does the player actually do here? If choices are central to the finished game, the opening needs a choice whose consequence can be felt. If the pleasure is close reading and atmosphere, it needs enough uninterrupted scene work to prove that rhythm. If routes are the main appeal, the player should at least see the hinge that will eventually separate them, even if the first public version does not contain a complete route.

Write one sentence for each promise before you cut anything. If you cannot, the problem is probably not length. The opening itself may not yet know what it is promising.

Why is the opening chapter not automatically the right stopping point?

The cleanest production choice is to export the first chapter and stop at its last line. Sometimes that is exactly right. Chapters often end on a turn, assets are already grouped around them, and the full game can continue without a special branch.

But chapter boundaries serve the reader who has already committed to the story. The stopping point of a first public version serves someone who is still deciding. Those jobs overlap, but they are not identical.

An opening chapter may spend most of its time establishing routine because the disruption lands in chapter two. That can be patient storytelling in the full work and a weak sample in isolation. It may end after a gentle transition because the next chapter was never meant to be separated by a long wait. Or it may introduce a choice one minute before the chapter break and stop before the reader sees why the choice mattered.

Do not rewrite the whole opening into a trailer. Instead, look for the smallest clean extension that lets the existing material complete its promise. That may mean including the first scene of chapter two. It may mean ending slightly before the chapter break, after the protagonist makes an irreversible decision. What you show first should be a deliberate edit of the opening, not an automatic file range.

Where should the cut actually go?

Look for a beat with two qualities: something has just changed, and the reader can imagine meaningful consequences that have not happened yet.

A good cut can follow the first result of a choice. The player selects whether to trust someone, the next scene shows that person opening one door and closing another, and then the public section ends. The consequence does not need to be large. It only needs to prove that the input entered the story rather than disappearing into a dialogue variation.

It can follow a revelation that reorganizes what the reader already saw. The strange detail in the opening gets an explanation, but that explanation creates a better question. It can follow a moment of commitment: the protagonist accepts the invitation, boards the train, agrees to the bargain, or decides to stay. Movement has begun, while the destination remains unrevealed.

What usually fails is stopping immediately before the change. A black screen after a loud noise is technically a cliffhanger, but if the opening has not yet delivered character, interaction, or a distinct tone, it feels like missing content rather than anticipation. Suspense works after satisfaction, not instead of it.

Also avoid ending on a settings screen, route menu, or development disclaimer. Put practical notes after the narrative endpoint, outside the story if possible. The last emotional memory should belong to the work, not its file status.

How much of the full story should you reveal?

Enough to establish the engine, not enough to resolve its main cycle.

For a mystery, reveal that clues can be interpreted and that an interpretation changes action. Do not solve the central case. For a romance, let attraction or conflict become specific and mutual. Do not spend the scene that settles whether the pair can be together. For a route-based drama, show one meaningful branch signal. Do not summarize all route premises as a selection catalog.

This is why percentages are unreliable. Ten percent of a two-hour kinetic novel and ten percent of a forty-hour route game do entirely different work. A word count is useful after you choose the cut, because it helps estimate reading time and production cost. It should not choose the cut for you.

The section you show first also needs to represent the finished game's density honestly. Do not move every piece of art into the first twenty minutes if the rest of the game is mostly portraits and backgrounds. Do not add choices every two minutes if the full version regularly reads for fifteen. The player is not only asking whether the opening is good. They are asking whether it represents the rest of the work.

Should the first public version contain multiple routes?

Usually it should expose the route structure without trying to complete a miniature version of every route.

Showing three short branches can look generous, but it triples the surfaces that need testing and often gives each relationship too little time to become real. A common route with one visible divergence is easier to understand. The reader sees that different futures exist, experiences one consequence, and leaves with a reason to wonder about the others.

There are exceptions. If route contrast is the entire concept, a compact scene that changes substantially under two choices may demonstrate the game better than a long shared opening. If the finished work is episodic, one self-contained case can work even when it is not literally the beginning. The test remains the same: does this section demonstrate the repeated pleasure of the full work?

Be careful about choices created only for the first public version and removed from the final game. They make feedback difficult to interpret and create extra work when that version has to be maintained. Whenever possible, cut from the real project and let the reader's path remain valid in the finished structure.

How do you test whether you have shown enough?

Give it to someone who has not heard your explanation.

Do not begin by asking whether they liked it. Ask them what they think the story is about, which character they expect to matter, what they believe their choice changed, and what they expect to happen next. Those answers map directly to the three promises. If they cannot answer one, adding ten minutes blindly will not help. Identify the missing scene function first.

Watch where attention changes. The moment someone starts tapping quickly, rereads a line, pauses before a choice, or looks surprised is more useful than a final duration. Ask where they first felt oriented and where they first wanted to continue. The distance between those moments is the active core of the opening. Material before it may need tightening; material after it may belong in the full release.

Run a second test after choosing the ending. A cut that looks elegant in the script can feel abrupt in play because music, transitions, and choice timing change the emotional shape. Test the last five minutes as carefully as the first five. Confirm that the public section clearly ends rather than crashes, that the player knows more exists, and that any link or next step works.

Steam notes that customers use demos to make purchase decisions and recommends that the quality match the real experience. Ren'Py's release guide similarly warns that script checking is not a substitute for thorough human testing. The section may be small, but it is not disposable. It often receives more cautious readers per minute than the full game ever will.

When should you publish it?

Publish when the represented slice is stable, not when the full game is stable.

The text inside the section you publish should no longer be changing structurally. Its art and sound should be close enough to final quality that they set honest expectations. Save behavior, text speed, rollback, audio controls, and the path out of the ending all need testing. A line promising that everything is temporary does not recover a reader's first impression.

Platform strategy comes after that. Steam supports demos as separate applications connected to the base game and allows developers to update them. Next Fest requires a publicly playable demo by the event, while itch.io can host a public, unlisted, password-protected, or restricted project depending on whether you are doing promotion or a closed test. Those are distribution choices. None repairs a slice that ends before it has shown why the story matters.

How does Novelez help with the first public slice?

Novelez is useful here because the first playable version can stay a story project rather than becoming a special technical build too early.

Choose the scenes that satisfy the three promises, keep the actual choices and assets attached to them, and publish that version to a link. A new reader can open it without installing a test package, and you can watch the same beginning they saw instead of trying to reconstruct their path from a script export. When a test shows that the endpoint lands too early, extend the project by one scene and publish the revised version again.

The important part is not that publishing becomes instant. It is that iteration stays close to the writing decision. You are testing whether a reader understands the promise, not whether you can maintain a separate preview codebase.

Once the cut survives real readers, that same structure can guide a later store demo or exported build. You have already answered the expensive question: what belongs in the sample?

What should be true before the first public section ends?

• The reader can name the central tension without reading a synopsis.
• At least one character has shown a specific private want or contradiction.
• The finished game's real interaction pattern has appeared.
• If choices matter, at least one choice has produced a visible response.
• The ending follows a change, revelation, or commitment rather than interrupting one.
• The sample's art, choice frequency, and reading rhythm honestly represent the full work.
• A first-time reader can explain what they expect next.
• The last screen makes it clear that this section is complete and the story continues.
• The entire slice, especially its last five minutes, has been tested as a player would experience it.

The first public section of a visual novel is not a percentage of a manuscript. It is a complete argument for continuing. Make the promise, let the player touch it, show one consequence, and stop when the next question is stronger than the one they arrived with.