Your Growth Plan Should Be A Set Of Decisions, Not A Document

Keith Sango

3 days ago

6 min read

Most growth plans are lists of aspirations with a budget attached. They’re comfortable to write, easy to approve and almost impossible to be wrong, which is the problem. A plan that can’t be wrong can’t be useful.

A useful plan forces at least one uncomfortable “no”. Richard Rumelt’s Good Strategy/Bad Strategy argues that real strategy has a kernel: a diagnosis of what is actually going on, a guiding policy for dealing with it, and coherent actions that carry the policy out. Most growth plans skip the diagnosis and jump straight to the actions.

A growth plan has five decisions, and if yours doesn’t make them, it’s a document.

Five decisions. Everything else in the plan is detail.

Start With The Constraint, Not The Goal

“Grow revenue by 50%” tells your team nothing about Monday. SMART goals help, but only once you work backwards to the inputs that drive the number, then ask what is actually limiting you.

New revenue comes from how many qualified leads you get, how many become customers, what each customer is worth and how long they stay. Those inputs multiply, so a 50% goal doesn’t need 50% from anywhere. Spread evenly across four inputs it’s roughly 11% each. That is a different and far more tractable conversation than “grow by 50%”, and it shows you where the cheapest 11% is hiding.

The output of any system is capped by its tightest bottleneck, so improving anything else is wasted effort. In growth terms, doubling lead volume into a sales team that is already at capacity doesn’t double revenue. It doubles the queue.

Teams often pour effort into the bottleneck they find most interesting. Founders who love content write more content, sales-led founders hire more reps, engineers rebuild the product. The plan’s first job is to name the real constraint, and the honest way to do that is to look at the data before you consult anyone’s preferences. Three questions get you most of the way:

Where does work pile up, and where do people leak out?
Which single input, if it doubled, would move revenue the most?
What would break first if demand doubled tomorrow?

The Four Growth Types Are A Risk Ladder

Market penetration, market development, product development and diversification are usually taught as four equal options. In practice each pillar changes more of what you know, so it helps to count the unknowns.

Market penetration: same product, same customer. Nothing new to learn, only more and better of what you already do.
Market development: same product, new customer. New buyers, new channels, new objections.
Product development: new product, same customer. You know the buyer, but not yet whether you can build, price and support the thing, and it borrows time from the team serving your core.
Diversification: new product, new customer. You are a beginner at everything at once.

The middle two pillars each add one big unknown, and which is riskier depends on whether your edge lives in the product or in the relationships. The outer two are not ambiguous.

Climb in order. Move up when penetration returns less for each pound spent, not when it gets boring.

Ambitious teams skip pillars. A company that hasn’t saturated its core gets excited about a new market or a second product line, and the venture drains the attention the core needed. Exhaust the cheap, well-understood growth first. Before you climb, ask three questions:

Is the next customer in your core getting more expensive, quarter after quarter, despite good execution? That is diminishing returns. Boredom is not.
Is the core healthy enough to survive the distraction? Stable retention and a delivery team that isn’t already stretched.
Can someone own the new pillar full time? A pillar added on top of everyone’s day job usually gets neither their best attention nor a fair test.

The ladder is a default, not a law. If your core market is shrinking, or one customer segment is turning out to be far better than the rest, market development may rightly come first. The point is to climb on purpose, with evidence and an owner.

Diversification spreads risk across revenue streams. It can also dilute the brand and strain integration, so go in phases and take advice on what you haven’t done before.

Be Specific About Your Customer And Your Edge

“SMEs” is not a target market. Segment by who they are, what they care about and how they behave, then build personas your marketing and sales people would actually recognise. Behaviour and trigger events, such as what made someone start looking this month, often tell you more than age or industry.

A good persona changes a decision: which channel, which headline, what to leave out. If you could swap in a different persona and make exactly the same choices, the persona isn’t doing any work.

The same goes for competitors. Don’t just list their features and pricing. Look for what they structurally can’t or won’t do, and build your value proposition there. The distinction matters:

Can’t comes from their size, cost base, legacy products or incentives. A large player can’t profitably serve very small accounts. A platform can’t be neutral about its own products. This kind of gap is durable.
Won’t comes from their brand or strategy. They choose not to customise, not to give expertise away, not to take smaller clients. This is a bet that they stay disciplined, so it can close the moment they change their mind.

Operations Break Before Strategy Does

A plan that adds demand without adding capacity just produces unhappy customers faster. Before you scale a channel, run a load test on paper. For each function (delivery, onboarding, support, finance, supply) write down how many extra customers a month it can absorb before service slips. The lowest number is your real growth ceiling this quarter, whatever the marketing target says.

Fix the process before you automate it, because automating a messy handoff only gives you a faster mess. Once it is stable, automate the repetitive work and put process reviews in the calendar so they actually happen. Sometimes the highest-return growth work is lifting the ceiling rather than adding more demand underneath it.

Sequence Around Your Cash

When money is tight, rank initiatives by payback speed and reversibility: cheap, quick, reversible bets first, large commitments once you have evidence. Amazon’s shareholder letters describe this as the difference between one-way and two-way doors. Most growth tests are two-way doors, so make them fast. A multi-year contract, a major hire or a rebrand is much closer to a one-way door and deserves evidence first.

Fast payback, reversible: do these now.
Fast payback, hard to reverse: proceed, but with a trial first.
Slow payback, reversible: fund a small, capped version.
Slow payback, hard to reverse: wait for evidence, or don’t do it.

Loans, crowdfunding and investors are options, and so are partnerships that bring resources without a cheque.

Write Down What You Won’t Do, And What Would Make You Stop

Every plan already has a “not doing” list. It’s just usually invisible, and decided by whoever argues loudest. Write it down and show it to the team. It is the most reliable evidence that the plan made a trade-off, and it protects the priorities from the idea that arrives in week three.

Then name your top few risks, attach a contingency to each, and decide in advance what result would make you stop. Pre-mortem is a good way to find the risks: imagine it is a year from now, the plan has failed, and write down why.

Make the case for setting kill criteria before you start, while you’re still objective, because once time and money are sunk we all judge badly. A kill criterion is specific enough to be uncomfortable, for example:

If cost per qualified lead is still above RX after eight weeks and three rounds of creative, we stop and reallocate the budget.

Teams with kill criteria avoid the sunk-cost trap.

Make The Plan Run

The plan is a loop. Reality updates it.

Communicate it so people can repeat the priorities in one sentence. Ask three people on different teams for the top priority, and if the answers differ, the plan hasn’t landed. Resource it properly, because priorities without people, budget and time are wishes. Then review KPIs on a fixed cadence:

Weekly: leading indicators and blockers, kept short.
Monthly: decisions. Continue, scale or kill.
Quarterly: re-plan. Has the constraint moved?

Look at what the data is telling you, not only whether you hit the number. A hit can hide a fragile process, and a miss can hide a good bet that needs another cycle. And expect the constraint to move. Fix lead flow and conversion becomes the limit, fix that and capacity is next. A plan that doesn’t expect its own bottleneck to shift will be stale within two quarters.

Keep a decision log: the date, the decision, the reasoning, what you expect, and what would change your mind. Six months on, it is the most valuable document you own, because it is the only honest record of what you knew at the time. When facts change, re-plan.

The Short Version

Choose the real constraint. Climb the risk ladder in order. Be specific about customer and edge. Make sure operations can carry the load. Sequence around cash. Write down what you won’t do and what would make you stop. Review often enough that the plan stays true.

One exercise for this week: open your current plan and underline every sentence a reasonable person could disagree with. If you’ve barely underlined anything, you have a document, and the decisions are still waiting to be made.