# Growth experiment

Designed for: Small teams trying to improve one point in a customer journey

Result: A focused growth test with a verified baseline, one primary metric, and draft materials ready for review.

## Bots

1. Researcher
2. PULSE
3. Writer

## Information to gather

- The growth goal and selected customer stage
- Approved data and metric definitions
- Channels, budget, capacity, privacy rules, and voice examples

## Workflow

### 1. Researcher

Check customer evidence and possible causes of the selected problem

Output: A concise evidence brief with gaps marked

Message to send:

Use the approved inputs I provide. Check customer evidence and possible causes of the selected problem. Return this result: A concise evidence brief with gaps marked. Ask me about missing information before you continue.

### 2. PULSE

Define the baseline and design one measurable experiment

Output: An experiment plan with one primary metric and success and stop rules

Message to send:

Start with this result from Researcher: A concise evidence brief with gaps marked. Define the baseline and design one measurable experiment. Return this result: An experiment plan with one primary metric and success and stop rules. Ask me about missing information before you continue.

### 3. Writer

Draft the approved test materials for the selected channel

Output: Two clear variants with claims and approval questions marked

Message to send:

Start with this result from PULSE: An experiment plan with one primary metric and success and stop rules. Draft the approved test materials for the selected channel. Return this result: Two clear variants with claims and approval questions marked. Ask me about missing information before you continue.

## Overall request

Design one growth experiment for this approved goal. Verify the baseline, use one primary metric, set success and stop rules before the test, and draft two variants without changing a live campaign.

## Decisions for a person

- Approve the problem interpretation, metric, and decision rules
- Approve targeting, budget, outreach, and any tracking change
- Approve the variants and decide whether to launch, revise, or stop

## First test

Use one approved historical data export to design a test and draft two variants without changing a live campaign.

## Hermes Desktop setup

1. Open each Bot's page in Bot Cabinet's Hermes Bots collection and follow its manual Hermes Desktop setup steps.
2. Run each step in that Bot's own chat and review the result.
3. Pass the approved result to the next Bot with the message provided for that step.
4. After the sequence works, you may create a group with the same Bots.
5. In a group, @mention the Bot you want. Membership order does not control who responds.