Planning profitable growth

0:00 Planning profitable growth

Every ecommerce founder has heard it: we need to grow. But grow what, and at what cost? Revenue can grow while profit shrinks. Customers can grow while cash runs out. Profitable growth planning brings everything in this course together into a plan that sets targets, allocates budget, forecasts scenarios, runs experiments and reviews results on a steady rhythm. In this lecture you'll learn to start from a target and work backwards, allocate budget, forecast with scenarios, use modern measurement like marketing mix modelling, and communicate the plan. Then you'll watch me model three scenarios in Python.

0:42 Why plan

Why plan at all? Because without a plan, budgets drift towards whatever shouted loudest last week: a new platform, a competitor's campaign, a big sale. A plan forces trade-offs into the open. It connects marketing spend to contribution and cash. It tells operations what stock and capacity they'll need. And it gives you a baseline to learn from, so next quarter's plan is better than this one.

1:11 Work backwards

Start from a target and work backwards. Say the goal is a certain contribution profit next year. How many orders does that need, given contribution per order? How many from returning customers, given your cohort repeat rates? How many new customers, then? At what acquisition cost, given realistic CAC by channel? And does the cash position support paying that CAC upfront? Working backwards turns an ambition into a set of numbers you can challenge.

1:43 Budget principles

Budget allocation principles. Protect the foundations first: product pages, checkout, speed and retention flows, because they improve every channel. Fund proven channels to the point where their marginal CAC is still acceptable, not their average. Reserve a test budget, perhaps ten to twenty percent, for new channels, creative and offers. And remember that returns diminish: the next pound in a channel usually buys less than the last one. That's why spreading across proven channels often beats pouring everything into one.

2:18 Scenario forecasting

Forecast with scenarios, not a single number. Build conservative, base and stretch scenarios, each with its own assumptions for new customers, CAC, contribution per order and repeat rates. Crucially, assume CAC rises as you scale, because it usually does. Add seasonality, with peaks for Ramadan, Eid, White Friday or Christmas. And list what would have to be true for each scenario. Then plan stock, cash and hiring around the base case, with a trigger for moving to stretch.

2:52 Measurement for planning

Modern measurement for planning. Marketing mix modelling estimates each channel's contribution from historical spend and sales, without tracking individuals. Open-source tools include Google's Meridian and Meta's Robyn, but they need a couple of years of weekly data and careful setup. Incrementality tests, like geo holdouts and conversion lift studies, calibrate those models and platform attribution. The practical approach is to triangulate: platform attribution for day-to-day optimisation, MER and contribution for monthly steering, and incrementality and MMM for budget allocation.

3:26 Simple example: Lahore fashion

A simple example. A small Lahore fashion brand wants to double revenue. Working backwards, doubling revenue at current CAC would need more cash than they have, because Instagram CAC rises as they spend more. So the plan focuses on levers that don't need proportional ad spend: a size guide and COD confirmations to lift conversion, a WhatsApp win-back flow for lapsed customers, and one new channel test on Daraz. Revenue growth is more modest than doubling, but profitable and fundable.

4:01 Realistic example: UK wellness (illustrative)

Now a realistic scenario with illustrative numbers. A UK wellness brand models three plans. Conservative: three thousand new customers at a CAC of thirty-two. Base: four and a half thousand at thirty. Stretch: six and a half thousand, but CAC rises to forty because they'd have to push into less efficient audiences. With first-order and twelve-month repeat contribution, and fixed costs of about ninety to ninety-five thousand, the base plan produces the highest profit after fixed costs. The stretch plan adds customers but not profit. The team chooses the base plan, with a trigger to scale if CAC comes in lower than expected.

4:46 Watch me do it: scenario planner

Watch me model the scenarios. I write a short Python function that takes new customers, CAC, first-order contribution, twelve-month repeat contribution and fixed costs. It calculates marketing spend as customers times CAC, then contribution from first and repeat orders minus marketing, then profit after fixed costs. I run three scenarios. Conservative gives a loss after fixed costs. Base gives the best profit. And stretch, with its higher CAC and fixed costs, lands below base despite having two thousand more customers. The numbers make the case clearly: more customers don't automatically mean more profit.

5:26 The rhythm

Governance: the weekly and monthly rhythm. Weekly: a short review of spend pacing, CAC by channel, stock levels and site health, with pre-agreed rules for pausing or scaling. Monthly: the growth review from the first lesson of this course, covering the five levers, contribution after marketing, cohorts and experiments. Quarterly: re-forecast the scenarios, reallocate budget based on incrementality evidence, and choose the next quarter's experiments. Rhythm beats heroics.

5:56 An experimentation roadmap

An experimentation roadmap belongs inside the growth plan. For each quarter, choose a small number of experiments tied to the limiting lever: a conversion test on product pages, a new retention flow with a holdout, a pricing or bundling test, a new channel test with a clear budget and success threshold, or an incrementality test on your biggest paid channel. Write down the hypothesis, the metric, the budget and the decision rule for each. At the end of the quarter, the plan gets updated with what you learned. That's how a growth plan becomes a learning system rather than a spreadsheet.

6:40 AI and responsible planning

Using AI in planning, with care. AI assistants can help build forecasts, clean data, draft plans and stress-test assumptions, for example by asking: what would have to be true for the stretch scenario? Keep every number traceable to your own data. Never let a model invent benchmarks or market sizes. And have a named person own each assumption in the final plan. Sustainability matters too: growth that depends on deep discounts, unsustainable returns, or misleading claims isn't growth you can keep.

7:15 Common mistakes

Common mistakes. Planning revenue instead of contribution. Assuming CAC stays flat as spend grows. One forecast instead of scenarios. Ignoring cash and stock timing. Allocating budget by last-click attribution alone. No test budget. And plans that are never reviewed after they're approved.

7:33 Communicating the plan

Finally, communicate the plan so people can act on it. One page for the whole team: the target, the limiting lever, the three or four priorities, the budget split, the key scenarios and triggers, and who owns what. Share the assumptions openly, because a plan people understand is a plan people can improve. Explain what operations need to prepare, like stock and courier capacity for peak seasons. And agree how you'll report progress each month. A plan that lives only in the founder's head can't guide anyone else.

8:11 Recap

Recap. Plan growth from a contribution target worked backwards. Protect foundations, fund channels to an acceptable marginal CAC, and reserve a test budget. Forecast conservative, base and stretch scenarios with rising CAC and seasonality. Triangulate measurement with platform data, MER, incrementality and MMM, and run a weekly, monthly and quarterly rhythm. Try this now: run the scenario planner from the lesson text with your own numbers, pick a base plan, and write down the trigger that would move you to stretch.