# Build an actor (/tutorials/first-actor)

# Build an actor that behaves like your audience [#build-an-actor-that-behaves-like-your-audience]

Setup writes a starting cast of actors from your site. This tutorial builds
one of your own from scratch: instructions, a personality and knowledge that
make it behave like a genuine member of your audience, then a first run to
check it does.

<PageFacts
  items={[
    { label: "Time", value: "15 minutes" },
    { label: "You need", value: "A project" },
    { label: "Next", value: "Write checklists" },
  ]}
/>

<Screenshot
  name="actor-create"
  alt="An actor's profile in Stunt Double: name, description, instructions and location."
  caption="An actor is named by role and described in plain language. The Knowledge tab holds what it knows."
/>

## Create the actor [#create-the-actor]

Open **Actors** in the sidebar and choose **Create actor**. Creating one takes
two steps:

1. **Start**: pick a template for a common type of user, or **Write your own**.
   A template fills in the next step rather than creating the actor straight
   away, so you see everything before it is saved.
2. **Details**: a **name**, an optional **description** and the
   **instructions**. Name the actor by role ("Busy shop owner"), never as a
   person, so results read clearly.

Everything here can be changed later on the actor's **Profile** tab.

## Start with a persona, not a checklist [#start-with-a-persona-not-a-checklist]

Instructions work best when they describe who the actor is standing in for,
not what you want it to click. A persona has a goal, a level of product
familiarity, and a tone:

> You are a small business owner evaluating whether to switch invoicing
> software. You are price-sensitive, comfortable with software but impatient
> with jargon, and you will abandon a signup flow if it asks for a credit card
> before showing you the product.

An actor written this way makes the same kind of judgment calls a real
visitor would, rather than mechanically executing steps. Write instructions to
the actor, in the second person. They hold up to 1,000 characters, so facts it
needs to look up belong in knowledge instead.

## Set its personality [#set-its-personality]

The **Personality** section of the Profile tab has three sliders:

- **Skepticism**, from trusting to skeptical: how much proof it wants before
  taking the product at its word.
- **Technical literacy**, from novice to expert: how comfortable it is with
  software, jargon and unfamiliar interfaces.
- **Frustration**, from patient to short fuse: how quickly a struggle wears it
  down, and how soon it gives up on a task.

The personality applies everywhere the actor works: chats, checklist runs and
interviews. Leave a trait unset to let the instructions decide. An interview
can override a trait for its own participants.

## Give it knowledge to draw on [#give-it-knowledge-to-draw-on]

Knowledge is the reference material an actor can pull from mid-run. Add
entries on the actor's **Knowledge** tab, each as one of four kinds:

| Kind | What it holds | Example |
| --- | --- | --- |
| Expertise | A skill or standard the actor applies | Accessibility auditing |
| Process | Steps the actor follows, in order | Reporting a bug |
| Reference | Facts to look up: an FAQ, a policy, pricing | Refund policy |
| Account | A test account to sign in with, never a real password | Staging test account |

The actor is never handed everything at once. It sees the titles of its
expertise and processes and opens the one a task needs before starting it,
and it searches reference and account entries before guessing a fact or a
login. Add an entry for anything you would otherwise have to explain to a new
team member on day one.

Keep entries short and specific, with clear titles. A dozen focused entries
work better than one long document, since the actor finds entries by what they
are about. An entry holds up to 8,000 characters; split anything longer into
entries with titles of their own.

## What an actor can do [#what-an-actor-can-do]

Out of the box an actor works through your product the way a real visitor
would:

- **Drives a real browser** through your product, observing each page and
  deciding what to do next. The **Browser** tab sets where that browser runs
  and where sites see it browsing from.
- **Reads email** sent to an inbox of its own, so it can pick up verification
  codes and magic links during sign-up. Each actor gets one address, assigned
  by the platform, on the Startup plan and above.
- **Stays signed in** between runs once you log in as the actor from the
  Browser tab, on the Startup plan and above.
- **Searches the web** in chat when it needs information beyond your product.

Email is the only channel an actor has. It never enters a phone number or
picks SMS verification; when a site insists, it reports the step as a blocker.

You don't have to wire any of this up. Give the actor a clear persona and the
knowledge it needs, and it draws on whatever is relevant to the step it's on.

## Test it before you rely on it [#test-it-before-you-rely-on-it]

Before building a whole checklist suite around a new actor, run one small
checklist against it and read the result. Confirm it interprets your
instructions the way you intended, then expand from there.

## Next steps [#next-steps]

- **[Write checklists that catch real regressions](/tutorials/effective-checklists)**
- **[Run a user interview](/tutorials/running-an-interview)**