The Growth Levers Process
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- —
A four-step operating process for finding the small number of actions that actually grow a business. You start by mapping a growth model — a mathematical representation of how the business acquires, engages, retains and monetises customers — so you can see where the points of highest leverage sit. Then you map the customer journey using jobs-to-be-done interviews, which has nothing to do with your product and everything to do with what the customer was already trying to do. You filter your idea list against the growth model, design experiments that test the biggest ideas in a week or two, and finally shift mindset: unlike school or a normal job, speed and mistakes are the point, provided every experiment and learning gets documented.
Origin
Extracted from Deep Dive with Ali Abdaal
How to run it
- 1
Map your growth model
Build a mathematical representation of how your business acquires, delights, engages, retains and monetises customers. Once real data sits in the model you can see mathematically where the points of highest leverage are, and which step is rate limiting.
Pro tip Keep asking to peel things back, consolidate and simplify — a model nobody can hold in their head won't get used.
- 2
Work out where growth actually sits in your organisation
Growth is cross-functional, not a department. In early-stage companies the founder should be head of growth, because growth requires understanding every aspect of the business and the authority to pull any lever — engineering, marketing, finance. In other companies the growth engine turned out to be compliance or even legal.
Pro tip When hiring into growth, weight relationship-building heavily: can this person get people in other departments to do things their boss didn't tell them to do?
Watch out Treating growth as a sub-function of sales and marketing hides levers that live in product, ops or compliance.
- 3
Map the customer journey with jobs-to-be-done interviews
Nobody woke up looking for your product, and often nobody woke up looking for your category. Work out what the customer was trying to do, what they thought they were looking for, where they were looking, and what questions they had — then turn up in that place looking like the thing they wanted.
Pro tip Ask where the person physically is at the exact moment they realise they have the problem, and advertise there.
- 4
Filter your ideas against the model
You will generate far more ideas than you can run. Filter first by asking which ideas hit the rate-limiting step in your growth model and therefore have the biggest impact on the business. Accept that most of the surviving ideas will still be bad.
Watch out Most ideas are bad, so filtering alone is not enough — the filter only decides what gets tested first.
- 5
Design and run growth experiments fast
For every idea you're excited about, work out the quickest way to test it. Big ideas — publishing a book, building a product, running a campaign — are hard and slow, so isolate the risky assumption inside them and find a test you can run in a week or two.
- 6
Make the mindset shift and document everything
Succeeding at school or in a job means doing everything asked, thoroughly, without mistakes. A startup is the opposite: you can't do all the things, you will make mistakes, and the mistakes carry the learning. Document each experiment and what you learned, or the same mistake gets repeated years later by you or someone else.
Pro tip Make talking about mistakes safe — people who fear a bad grade or getting fired won't report them, so nobody learns.
Watch out Without written learnings you rediscover the same lesson two years later.
In the wild
Anyone using a fintech has to pass KYC — verifying documents, source of funds, that you're not a fraudster. For most companies that's a giant pain, a box to tick, and the head of compliance's goal is simply to be 100% compliant. Wise reframed the job: get as many weekly active users and as much money through the platform as possible while remaining compliant. That changed how the compliance team thought about their work. They smoothed the process out and spent serious time optimising the user journey so people could get up and running quickly and painlessly. Their compliance function became a component of their growth engine.
→ A function normally treated as friction became one of the company's biggest growth levers.
It didn't take a genius to work out that people would rather stream music than buy an album to listen to one song. The problem was the US record labels didn't want it to happen. So Spotify started in Europe, entering countries with more favourable laws. They built momentum, audience, money and eventually enough lawyers to move into the US. For a period, the function driving growth wasn't marketing or product — it was legal. The lesson is that the growth lever can sit anywhere in the organisation, so you have to work out from first principles how your business grows and then align whatever organisational pieces are needed.
→ Regulatory strategy, not marketing spend, unlocked the largest market.
Common mistakes
Boxing growth into a department
Growth is so cross-functional that putting it in a department creates a problem even in large organisations. If the growth lead can't tell an engineer, a marketer and the finance person what to do, they can't pull the levers that matter. In early-stage companies that person is the founder unless there is a very good reason otherwise.
Testing the whole idea instead of one assumption
Big ideas contain many risky assumptions, and the only way to test them all at once is to actually build the thing — three and a half years for a book, months for a product. That is not an experiment, it's a bet. Isolate one assumption and test that.
Running experiments without writing anything down
Teams routinely run something, forget it, and only remember two years later after making the same mistake again. Documenting the experiment and the learning is part of the process, not admin.
From the transcript
“there's four steps to this process um the first thing which I think we're going to do today is you sort of map your growth…”
“step two is you need to really understand your customer's Journey which has nothing at all to do with your product nobody if you're a…”
“so the next thing you got to do is experiment quickly so step three is to design and run growth experiments as quickly as possible”
From the episode
How I’m Scaling My Business from 7 to 8 Figures