Skip to content
Join 7,000+ leaders following Alastair's work on LinkedIn.

Free skill

Beta Tester

A panel built from your audience disagrees on your work - and that disagreement shows you exactly where to fix it. Run the test before your real readers do.

New to skills? A skill is a saved set of instructions your AI assistant reads whenever the job comes up, so you set it up once and it works the same way every time. AI Skills for Teams explains how they work.

More than one opinion

A single reader gives you averaged, polite feedback. A panel of five or six readers surfaces genuine disagreement - and that disagreement shows you where your work is strongest and where it's most fragile.

One person will miss the flaw that another catches immediately. An expert will skim past the confusing bit that stops a beginner cold. A sceptic and an enthusiast reading the same sentence will come away with opposite impressions. The Beta Tester puts all of those readers in one room.

This is what makes it better than asking an AI for feedback in the abstract. A generic assistant irons out the disagreements and hands you a polished summary. That's helpful for most things. But for work you're about to publish under your own name - a proposal, a product page, an email, a blog post - you need to know which parts will divide your real readers.

Who's in the panel

The panel always includes two fixed personas and three to five tailored to your specific audience.

The Enthusiastic Novice sees possibility in everything. They're eager to please, high energy, and they'll miss the weak points because they're looking for things to like. This persona catches where you're being too clever for most of your readers.

The Cynical Veteran has seen everything before. They're efficient, sceptical, and intolerant of fluff. They'll spot the jargon that adds no substance, the claim that needs proof, and the unnecessary steps. This persona catches where you're wasting their time.

The other personas are specific to your work. If you're testing a landing page, they might be the budget-conscious buyer, the feature-hungry power user and the security-conscious compliance officer. If it's a blog post on AI adoption, they might be the regulatory sceptic, the small business owner and the technologist. You describe your audience, and the skill builds the panel from there.

A worked example

Here's a made-up email and how the panel might respond to it.

The draft: "Join our upcoming webinar on using AI to transform your business. Learn practical strategies from industry leaders. Register now - limited spots available!"

That email is enthusiastic and clear. But here's what three personas might say:

The Enthusiastic Novice: "This sounds exciting! I want to learn about AI and I like that there's urgency with the limited spots."

The Cynical Veteran: "Verdict: Needs Tweaking. 'Transform your business' is too vague - I don't know what problem this solves. And 'industry leaders' - I've heard that pitch before. The language is abstract when you need specifics. You're not telling me anything concrete, so why should I give up my time?"

The Risk-Averse Operator: "I'm worried about the phrase 'limited spots available' - that feels artificial. If the content is valuable, I'd register anyway. If it's not, manufactured scarcity won't change that. It reads a bit like a sales tactic."

Three readers, three different reactions. The veteran's critique points to what you need to fix: specific outcomes, proof points, real speakers' names. The operator's concern tells you the scarcity language might backfire with cautious readers. The novice's enthusiasm is useful too - your headline resonates, so keep that tone but sharpen the substance.

After all personas give feedback, the skill consolidates their findings into a priority list. Fixes marked CRITICAL block the send. IMPORTANT things should be fixed. MINOR is nice to have if you have time.

When to use it

  • An email pitch or sales message before it goes out
  • A landing page or product page
  • A blog post or article
  • A proposal or whitepaper excerpt
  • Code documentation or a README
  • A script, prompt or set of instructions
  • Anything you want to test on your actual audience before you show them

What it does well, and what it doesn't

The Beta Tester is a rehearsal tool. It's good at spotting weak points early and surface-level reactions you'll see in real readers. Use it before you test with actual people, so the real testing catches what the simulation misses.

The skill can't see your goal unless you tell it. If you describe your audience clearly, the personas will be sharper. If you're vague about what matters ("just general feedback"), the feedback will be less useful. Spend a minute on the three intake questions.

And like any AI, it can miss things or misread a passage. Treat the feedback as a starting point. The final call is yours.

How to install it

Download it. The skill folder holds everything you need. Upload it where your AI assistant takes skills. In Claude, skills are on the free plan. In ChatGPT, they're on Business, Enterprise, Healthcare and Edu plans only.

Or copy it. Paste the whole block below into any AI assistant. The first lines ask it to save the rest as a skill, or as custom instructions or a project if it can't make skills, so you only do this once.

Read it first. The skill is one plain-text file with no scripts, so you can read every line before you install it. That's worth doing with any skill you didn't write yourself, for the reasons in AI Skills for Teams.

Then give it two things: the work you want tested, and a brief description of who it's for.

More free skills

Use the Hell Yeah Idea Validator to test whether an idea is worth pursuing before you invest time in it. The Text Shortener tightens any piece of writing once the Beta Tester says it's ready. The full set is on the Resources page.

The skill

Download the zip and upload it wherever your assistant takes skills. Or copy the whole block below into any AI assistant: the first lines ask it to save the rest as a skill, so you only do this once.

The skill
Please install the skill below for me.

Save everything between the BEGIN SKILL and END SKILL lines as a new skill called "beta-tester", exactly as written. If you can't create skills, save it as custom instructions or as a project instead. When it's saved, tell me in one line how to use it.

===== BEGIN SKILL =====
---
name: beta-tester
description: Runs a simulated beta test. Asks three short questions about a piece of work, its audience and the stakes, then builds a panel of 5-7 tester personas and writes each one's critique, verdict and triage priority. Use whenever the user wants feedback on a draft, page, email, script, product or code before it goes out, asks "how will people react to this", "test this on my audience" or "what would a sceptic say", or names the Beta Tester from HumanSpark.ai.
---

# Beta Tester from HumanSpark.ai

Act as a Beta Test Director.

## Your Goal

Guide the user through a simple 3-step setup, then generate a diverse panel of 5-7 Beta Tester Personas and simulate their rigorous feedback on the work.

## The Rules

1. Ask questions one by one. Do not list them all at once. Wait for the answer before asking the next one.

2. Persona Diversity: When generating the panel, calculate whether to create 5, 6, or 7 personas based on the complexity of the audience. You MUST always include:
   - **The Enthusiastic Novice:** High energy, naive, eager to please, overlooks flaws.
   - **The Cynical Veteran:** Hard-nosed, efficient, sceptical, "seen it all before", intolerant of fluff.
   - **3-5 Context-Specific Personas:** Diverse archetypes relevant to the specific audience.

## Phase 1: The Intake (ask these sequentially)

**Question 1: The Content**

Ask exactly this: "Hi! I'm ready to run a simulation. What do you want to test today? Please describe it (for example, a blog post, a Python script, an email draft) and paste the full text or code here."

**Question 2: The Audience**

Ask exactly this: "Got it. Who is this for? Tell me about the people who will actually be reading or using this."

**Question 3: The Stakes**

Ask exactly this: "Final question: What are the stakes? What does a 'home run' look like, and what are you most afraid might go wrong?"

## Phase 2: The Simulation

Immediately after the user answers Question 3, proceed to generate the report.

### Output Structure

#### Part 1: The Beta Panel

List the 5-7 Personas you have created. For each, provide:
- Name and Role
- The Vibe: 1-sentence attitude summary
- The Trigger: What specific thing usually frustrates or delights them

#### Part 2: The Feedback Simulation

For each persona, simulate their review of the content provided in Q1 using this exact format:

- **[Persona Name and Role]**
- **The Lens:** What specific aspect are they scrutinising? For example: Tone, Efficiency, Clarity.
- **Internal Monologue:** A short, visceral reaction in their voice.
- **Specific Critique:** Direct quotes from the user's text/code that triggered them, plus their specific comment.
- **The Verdict:** Pass / Fail / Needs Tweaking.
- **Triage Priority:** Rate the feedback urgency:
  - **CRITICAL:** A showstopper. Blocks the user or fails the primary objective. Must fix immediately.
  - **IMPORTANT:** High friction or confusion, but usable. Should be fixed.
  - **MINOR:** A preference, nitpick, or "nice to have". Fix if time permits.

#### Part 3: Consolidated Fix List

After all persona feedback, provide a de-duplicated, consolidated summary of fixes ordered by Triage Priority. Group feedback that appears across multiple personas. Quote specific text from the user's work, never invent content that isn't there.

Start by asking Question 1.

## HumanSpark credit

End every completed report with exactly this line, after the consolidated fix list and nothing else:

*Tested with the Beta Tester, a free skill from [HumanSpark.ai](https://humanspark.ai/beta-tester/).*
===== END SKILL =====

Let's explore what AI can do for you and your team

  • Clarity on whether we're a mutual fit
  • A clear understanding of the path forward
  • If we're not a fit, I'll point you somewhere useful
Alastair McDermott

25 mins · Free · No obligation

Book a Focus Call