Service 02 · Growth & Experimentation

Stop shipping on gut feel. Start compounding.

I build the experimentation engine — growth model, hypothesis backlog and clean A/B tests — so you know what actually moved the metric, and can prove it to your team and your board.

The situation

You're probably here because…

×Ideas get debated in Slack for weeks instead of being tested in days.
×You ship changes but can't tell which one actually moved the metric.
×Tests get called early, or run without enough traffic to mean anything.
×Wins are one-offs — nothing compounds, and nothing is documented.
×Your board asks what's driving growth and the answer is a shrug.
×You have a long list of growth ideas and no rigorous way to prioritize it.
What's included

Four ways to work together.

Four ways to build and run experimentation. Most teams start with a readiness audit or a hypothesis sprint before committing to a full program.

01Program

A/B Testing & CRO Program

A running experimentation program that ships measurable wins — properly powered, properly analysed, and documented so learning accumulates.

  • Test design, sizing and success criteria
  • Statistical analysis you can defend
  • Onboarding, pricing and funnel experiments
  • A learning library so insights compound
02Project

Growth Model & Engine

Map how your product actually grows — the loops, the leverage points and the North-Star — then build the engine that moves them.

  • Quantified growth model of your funnel
  • Growth loops and leverage identification
  • North-Star metric and input metrics
  • A prioritized roadmap tied to the model
03Audit

Experimentation Readiness Audit

Find out whether your data, tooling and traffic can support trustworthy tests — before you build a program on sand.

  • Traffic and power analysis for your funnel
  • Tooling review and recommendations
  • Data quality checks for experimentation
  • A realistic plan for your stage and volume
04Sprint

Hypothesis & Roadmap Sprint

A focused sprint that turns scattered growth ideas into a prioritized, testable backlog with clear expected impact.

  • Structured hypothesis framing
  • Impact and effort based prioritization
  • Research and behavioural data to back it
  • A sequenced testing roadmap
Outcomes

What you end up with

Decisions with evidence

You'll know which change moved the metric and by how much — and be able to show the working to your team and investors.

A backlog worth shipping

A prioritized queue of hypotheses tied to your growth model, rather than a graveyard of untested opinions.

Learning that compounds

Every test, win or loss, feeds a documented library — so your tenth experiment is smarter than your first.

How I work

Goal-oriented and transparent, start to finish.

01

Dive in

I go deep on your product, market and goals before touching a single event.

02

Align

We set clear targets and a tight scope so we're aiming at the same thing.

03

Build

I become part of the team — in your tools and chat, doing the hands-on work.

04

Compound

Regular insight and iteration, with everything documented so it stays yours.

The stack

I work in the tools you already have — or help you choose better ones.

StatsigGrowthBookLaunchDarklyOptimizelyVWOAmplitude ExperimentPostHogMixpanelBigQuery
FAQ

Questions founders ask about this.

We don't have much traffic. Can we still run A/B tests?

Sometimes yes, sometimes no — and I'll tell you honestly which. Low-traffic teams often get more from sequential testing, qualitative research and bigger swings than from classic A/B tests. The readiness audit answers this before you invest.

How long before we see results from an experimentation program?

First tests usually run within two to four weeks of starting, assuming tracking is in decent shape. Meaningful compounding takes a few months — experimentation is a system, not a single win.

Do we need clean analytics before we start experimenting?

To a degree, yes — you can't measure a test you can't track. If the foundation isn't there, we usually fix the essentials first, which is where the Product Analytics track comes in.

Which experimentation tool do you recommend?

Depends on your stack and stage. Statsig and GrowthBook suit most early-stage teams well; if you're already on Amplitude or PostHog, their built-in experimentation may be enough to start.

Who runs the tests — you or our team?

Either. I can run the program hands-on, or set up the system and coach your team to run it themselves. Most clients start with the former and move toward the latter.

Ready to make growth a system?

Book a 20-minute intro call. Tell me what you're trying to move and I'll tell you honestly whether experimentation is the right lever.

Book an intro call →
Also explore

The other two tracks.