“Grow the business” is not a brief you can execute — it's a wish. The job of an operator is to turn that wish into levers: a small number of variables you can actually move, diagnose and act on, one at a time.
The one formula worth memorizing
Everything a growth team does lands somewhere in this equation. A campaign changes traffic; a product-page redesign changes conversion; a bundle changes average order value; a retention flow changes repeat purchase. If an action doesn't move one of the four, it isn't a growth action.
The four levers, and the question each asks
Traffic — who is arriving, and why?
Traffic is the volume and quality of visitors across paid, organic, social and direct. The diagnostic question isn't only “how many?” but “are these the people who would buy?” More of the wrong visitors can make every other number look worse.
Conversion — why do visitors leave without buying?
Conversion is the share of visitors who place an order. The leaks live in specific places: offer clarity, product pages, trust, checkout friction. “Low conversion” is a symptom; the job is finding the page or step where intent is lost.
AOV — why don't buyers spend a little more?
Average order value is the worth of each order. Bundles, thresholds, upsells and the structure of the range all move it — but only when the suggestion fits what the customer already came for.
Repeat — why don't customers come back?
Repeat purchase is how often past buyers return. It's usually the cheapest growth available, and the most neglected — because post-purchase experience gets far less attention than acquisition.
The formula multiplies. One weak lever drags the whole result — no amount of traffic fixes a store that doesn't convert, and no conversion work helps a product no one returns to.
Don't pull every lever at once
The instinct under pressure is to work on everything: run a sale, start ads, redesign a page and launch emails in the same week. That's the fastest way to learn nothing. When several levers move together, you can't tell which action did anything — and you can't repeat what worked.
A disciplined loop runs differently:
- Diagnose to find the binding constraint. Look at where the equation is weakest relative to comparable brands and to your own history — not at which tactic is fashionable.
- Pull one lever. Run the actions that address that constraint, with the goal set by you.
- Attribute, then move on. Read what changed, keep what worked, and only then shift to the next lever.
Who sets the numbers? You do. A tool can show where the equation is weak and which actions address it — it shouldn't invent a target for you, and it can't promise a result. The target is a business decision; the evidence is what software provides.
Why this matters for a small team
A large team can work the levers in parallel. A small team can't — and shouldn't try. Decomposing the formula and sequencing one lever at a time is how a single operator gets the focus of a full growth team: fewer simultaneous bets, clearer cause and effect, and no afternoons spent on actions that move nothing.
Key takeaways
- Every growth action moves one of four levers: traffic, conversion, AOV or repeat.
- The levers multiply — one binding constraint holds back the whole result.
- Diagnose the weakest lever from data, not from whichever tactic is trending.
- Pull one lever at a time so you can attribute what actually worked.