Grow on evidence. Not on guesses.

Magic Improve studies your product data, tells you which experiments to run next, and remembers everything you learn.

PromoUp to 50% off your first year

Know what to test next. And why.

What Magic Improve shows you, run by run.

Your Product

Magic Improve · active

Mon 08:00 1. It studies your numbers

This week’s analysis

A fresh look at your data and everything you already know.

Analyzing Done

What it read

  1. 08:00 Product analytics, last 30 days
  2. 08:01 14 past A/B tests and their results
  3. 08:01 Your goals, notes and decisions
  4. 08:02 Your code, for context
  5. 08:04 Analysis done. 4 new ideas

Goal: more paying customers

North star

Trial to paid

4.2%

8 weeks agoThis week

Guardrails RefundsSupport tickets

Top idea Cut onboarding from 5 steps to 3 ICE 8.0

Mon 08:04 2. It ranks what to test next

4 new opportunities

Ranked by impact, confidence and effort.

  1. 1

    ExperimentMore paying customers

    Cut onboarding from 5 steps to 3

    ICE 8.0 · Impact 4 · Confidence 4 · Effort 2

  2. 2

    MeasurementMore paying customers

    Track why trials are cancelled

    ICE 7.5 · Impact 3 · Confidence 5 · Effort 2

  3. 3

    ExperimentHigher revenue per customer

    Show yearly savings on the pricing page

    ICE 6.0 · Impact 4 · Confidence 3 · Effort 2

  4. 4

    InvestigationMore active teams

    Find where invited teammates drop off

    ICE 4.0 · Impact 4 · Confidence 3 · Effort 3

Experiment More paying customers · Trial to paid

Cut onboarding from 5 steps to 3

8.0

Most new teams stop at step 3 of onboarding. Teams that finish it are far more likely to pay.

Deliverable
An experiment spec: variants, audience, success metric
Decision it enables
Whether every new team gets the shorter onboarding
Evidence
Signup funnel2 past testsYour notes
Keep Dismiss Mark testing

AI-generated · your team decides what runs

Mon 09:15 3. Your agent builds it

Memory

Carried into every run

  • Learning

    A yearly plan banner at checkout lost to the original.

    From a past A/B test

  • Decision

    No pricing changes until January.

    Logged by your team

  • Team input

    Enterprise trials are handled by sales.

    Note

Ideas you already tried won’t come back.

Claude Code your-product

> Build the top experiment from Magic Improve

magic-improve · Read brief

Cut onboarding from 5 steps to 3 · ICE 8.0

magic-improve · Read memory

1 learning and 1 decision apply

Adding a 3-step variant behind your feature flag

✓ Variant ready for your team to review

magic-improve · Log decision

Marked as testing

What you get.

A growth team’s judgment, on every run.

  • The next experiments worth running, ranked

    Scored by impact, confidence and effort.

  • Every idea tied to a goal

    One north star metric per goal. No vanity wins.

  • Evidence you can check

    Each recommendation shows the data behind it.

  • A memory that compounds

    Decisions and learnings carry into every run.

  • A brief your team can build from

    Or hand it straight to your coding agent.

  • Your team stays in control

    It proposes. You decide what runs.

Recommendations are AI-generated. Your team decides what to run.

Plugs into the tools you already use.

It reads your data where it already lives.

  1. 1 Connect your analytics
  2. 2 Set your goals
  3. 3 Review what to test

Product analytics

Where your numbers come from

Experiments

Past tests and their results

Your coding agents

Connected over MCP

Updates

New ideas, where you work

Optional: read-only access to your code, so ideas fit how your product is built.

MCP

Hand the brief to your coding agent

Your agent reads what to test, builds it in your code, and logs the decision back.

Magic Improve

Experiment brief

Cut onboarding from 5 steps to 3

  • VariantsCurrent, 3 steps
  • MetricTrial to paid
  • GuardrailSupport tickets
  • Claude Code
  • Codex
  • Cursor

Works with any MCP client.

  • Read ranked ideas and their briefs
  • Build an experiment from its brief
  • Log decisions and learnings
  • Start a fresh analysis

Start free. Grow on evidence.

Monthly prices in euro.

Join the waitlist: up to 50% off your first year.

On paid plans.

  • Free

    Try it on one product

    €0

    • One product
    • A few runs per month
    • Top opportunities
    • Updates by email

  • Startup

    Recommended

    For teams testing every week

    €99 /month

    • Unlimited runs on your product
    • Ranked opportunities with briefs
    • Memory of goals, decisions and experiments
    • MCP for your agents
    • Updates in Slack, Teams or email

  • Custom

    For several products

    Custom

    • Multiple products
    • Custom integrations
    • Single sign-on (SSO)
    • Volume pricing

Promo confirmed at setup.

Questions

Something we didn’t cover? Write to hello@magic-team.com.

Does it change my product?

No. It doesn’t write code or open pull requests. It proposes; your team or your coding agents build.

What does it need access to?

Your product analytics and your A/B testing tool. Access to your code is optional and read-only.

How is it different from my analytics tool?

Your analytics shows what happened. Magic Improve tells you what to try next, and why.

How does it learn?

Every decision, result and note goes into its memory, and into the next analysis.

Can it be wrong?

Yes. It’s AI. Every idea shows its evidence and confidence, and your team decides what runs.

How does the promo work?

Up to 50% off your first year on a paid plan, confirmed at setup.

Find your next win.

Up to 50% off your first year.