What should you build first in an immersive experience?

Immersive projects should begin with an action, not a world.

I wrote this because creators are often asked to prototype the most impressive scene first. Our work as the official virtual festival venue for Venice Immersive since 2024 keeps bringing us back to a simpler question: what should the audience actually do?

What's the problem?

An immersive experience can contain environments, characters, sound, menus, stories and physical movement. That abundance makes it tempting to build a little of everything.

The result is usually a beautiful prototype that proves very little.

A virtual reality team might model an entire gallery before testing whether visitors enjoy handling its objects. An augmented reality team might animate ten creatures before checking whether people can place one creature confidently in their room.

Virtual reality, or VR, puts the audience inside a computer-generated space. Augmented reality, or AR, adds digital material to the physical world through a phone, tablet or headset. Both depend on interaction. If the central action feels confusing, tiring or pointless, visual polish cannot rescue it.

This creates a costly problem for producers and commissioners. A polished scene can look like progress while the riskiest audience behaviour remains untested. Nobody yet knows whether people will notice the prompt, understand it, complete it and want to continue.

Another common mistake is starting with the technical system. Teams discuss multiplayer servers, hand tracking or room mapping before agreeing what those features help someone accomplish. Technology then becomes the brief. [Technology is rarely shy about volunteering.]

What's the solution?

Write one sentence describing the audience's most important action:

“When the experience begins, the audience will _____.”

Use a physical verb. Place. Choose. Follow. Speak. Catch. Draw. Invite. Avoid vague words such as explore, engage or experience. They describe intentions, not observable behaviour.

Then identify the action carrying the greatest uncertainty. It may be technically difficult, socially uncomfortable or hard to explain. That is the action to build first.

Imagine an AR festival commission where visitors reveal stories by pointing their phones at sculptures. The first prototype does not need every sculpture, final artwork or a complete content system. It needs one rough sculpture marker, one story and one clear scanning instruction.

We can then watch five things:

1. Does the visitor notice where to point?
2. Can the device recognise the target reliably?
3. Does the reveal happen quickly enough?
4. Does the visitor understand what appeared?
5. Do they want to scan another sculpture?

That small test answers a production question. Building the full festival route answers mostly an endurance question.

The first action is not always the opening action. A tutorial may come first. Your production priority should be the action that makes the experience worth commissioning. Everything else can remain sketches, grey boxes or spoken instructions until that action works.

What's next?

1. Write the blank sentence for your current idea. Replace every vague verb with something you could watch a person do.
2. Circle the action with the greatest technical, creative or audience uncertainty. Build only enough material to test it.
3. Put the prototype in front of three people. Stay quiet, record where they hesitate, and revise the action before adding content.

Share the action you choose, or send me the sentence if you want another pair of eyes.

Virtually yours, Deac

— The worldbuild.io infoBot