E-commerce Marketing and GrowthStore fundamentals and product pages · Lesson 1 of 20
The e-commerce growth model
Video lecture
The e-commerce growth model
The narrated lecture is in production
Every chapter is scripted and ready. Browse the chapters and read the full transcript now — the video will appear here when it’s published.
Chapters
Transcript of the narration, chapter by chapter.
0:00 The ecommerce growth model
Two online stores each make a hundred thousand in revenue this month. One is thriving. The other is quietly running out of cash. Same revenue, very different businesses. In this lecture you'll learn the ecommerce growth model: the handful of levers that produce revenue and profit, how to find the one that's limiting your store right now, how the channel mix has changed in twenty twenty-six, and a monthly growth review you can run in an hour. Then you'll watch me model what-if scenarios in a few lines of Python.
0:39 Why a model
Why start with a model? Because ecommerce advice is endless: post more reels, run more ads, launch a loyalty programme, redesign the site. Without a model, you can't tell which advice applies to you. A simple growth equation turns a vague goal, like grow the business, into specific questions: do we need more visitors, a higher conversion rate, bigger baskets, more repeat purchases, or better margins? Each answer points to different work.
1:10 The five levers
Here's the core equation. Revenue equals visitors times conversion rate times average order value. Then add time: repeat purchase rate turns one order into several. And add profit: margin, after product costs, shipping, payment fees, returns and marketing. So there are five levers. Traffic. Conversion. Order value. Retention. And margin. Think of it like a car. Traffic is fuel, conversion is the engine, order value is the gearbox, retention is the mileage per tank, and margin is whether the journey makes money.
1:46 Find the limiting lever
How do you find the limiting lever? Compare each one with your own history and, carefully, with reasonable expectations for your category. Is traffic healthy but conversion weak? That's a site, offer or trust problem. Is conversion fine but order value low? That's merchandising. Are first orders strong but few customers return? That's retention and experience. Are sales growing but cash shrinking? That's margin and unit economics. Fix the tightest constraint first, because improving any other lever won't help as much.
2:21 The 2026 channel mix
The channel mix has changed. Search is fragmenting: shoppers ask AI assistants and AI search features for recommendations, alongside Google Shopping, marketplaces and TikTok search. Clean product data now feeds all of them. Retail media, meaning ads sold by retailers and marketplaces like Amazon, noon and Daraz, is a major budget line. Automated campaigns, like Performance Max and Meta's Advantage+ sales campaigns, depend on your product feed, conversion data and creative variety. And social commerce keeps growing where it's available, which varies a lot by country.
2:58 Regional realities
Regional realities matter. In Pakistan, cash on delivery is still a big share of orders, which affects refusals, cash flow and fraud. In the Gulf, marketplaces like noon and Amazon are major channels, buy-now-pay-later is widely used, and Ramadan, Eid and White Friday dominate the calendar. In the UK and US, wallets and fast delivery are expected, and consumer rules on pricing and reviews are strict. The same five levers apply everywhere, but the constraints look different.
3:31 Simple example: Lahore clothing store
A simple example. A Lahore clothing store has plenty of Instagram traffic, but conversion is low. The team's instinct is to spend more on ads. The growth model says no: traffic isn't the constraint. Recordings and support chats show people unsure about sizing and about whether cash on delivery is available. Adding a size guide and showing COD availability on product pages lifts conversion, which makes every future ad more profitable. The model stopped them pouring more fuel into a car with a slipping engine.
4:08 Realistic example: UK homeware (illustrative)
Now a realistic scenario with illustrative numbers. A UK homeware brand gets sixty thousand sessions a month, converts one point eight percent, and has an average order value of forty-eight pounds. That's about a thousand and eighty orders and fifty-two thousand in revenue. Gross margin is sixty percent, variable costs per order are nine pounds, and marketing is fourteen thousand. Contribution after marketing is about seven thousand four hundred pounds. Which lever is worth pulling? The next scene runs the numbers.
4:43 Watch me do it: what-if model
Watch me model it. I write a tiny Python class with sessions, conversion rate, order value, gross margin, variable cost per order and marketing spend. Three methods: orders, revenue and contribution. Then three scenarios. Ten percent more sessions, but that costs ten percent more marketing. Ten percent better conversion. And ten percent higher order value from bundles. Running it, the sessions scenario adds revenue but only a little contribution, because the extra orders cost ad money and shipping. The conversion scenario adds more. And the order value scenario adds the most contribution, because bigger baskets don't add extra shipments. Your numbers will differ. The point is to see it before you spend.
5:32 Reading the results (illustrative)
Let me read the results out properly, because the numbers teach the lesson. The base month makes about seven thousand four hundred pounds of contribution. Ten percent more sessions, with ten percent more ad spend, lifts contribution to about eight thousand one hundred. Ten percent better conversion lifts it to about nine and a half thousand, because the same traffic produces more orders without extra ad spend. And ten percent higher order value lifts it to about ten and a half thousand, because each extra pound of basket carries no extra shipping or payment fee per order. Revenue is identical in all three scenarios. Contribution is not. That's why the growth model is judged by margin, not by revenue.
6:24 The monthly growth review
Run a monthly growth review, about an hour. Look at each lever: sessions by channel, conversion rate by device and source, average order value, new versus returning revenue, and contribution margin after marketing. Note what changed and why. Pick the tightest constraint. Choose one or two actions for next month, with an owner and a measure. And before spending more on ads, ask: is our conversion rate healthy, is the first order profitable or recovered by repeat purchases, and do we have stock and operations to handle growth?
7:02 Retention in the model
Retention deserves a special mention in the model, because it changes everything downstream. If a customer buys once, you have one chance to cover your acquisition cost. If they buy three times in a year, the second and third orders carry no acquisition cost at all, so they're often your most profitable revenue. That's why two stores with the same first-order numbers can have completely different futures. So add two numbers to your monthly review: the share of revenue from returning customers, and the percentage of last quarter's new customers who have bought again. If both are falling while ad spend rises, you're renting growth, not building it.
7:49 Common mistakes
Common mistakes. Chasing revenue instead of contribution margin. Spending more on ads when conversion or retention is the real constraint. Comparing yourself with generic benchmarks from different categories. Ignoring returns and cash on delivery refusals in the numbers. And treating every channel's reported revenue as if it adds up, when platforms often claim the same sale.
8:13 Recap
Recap. Revenue is traffic times conversion times order value, extended by retention and judged by margin. Find the tightest constraint, fix it first, and model what-ifs before you spend. Remember that the channel mix now includes AI-assisted search, retail media, automated campaigns and social commerce, all fed by good product data, and that regional realities shape every lever. Try this now: copy the growth model from the lesson text, enter your own last month's numbers, and run three ten percent scenarios to find your most valuable lever.
Revenue is a product of levers
E-commerce revenue can be broken into a small number of multiplying levers:
Revenue = Visitors x Conversion rate x Average order value (AOV)
Over a customer's lifetime:
Customer value = AOV x Purchase frequency x Customer lifespanAnd profit depends on what it costs to acquire and serve those customers:
Profit = Revenue - Cost of goods - Fulfilment & shipping - Payment fees
- Returns - Marketing - Operating costsGrowth comes from improving one or more levers without damaging the others. A discount can raise conversion rate but lower AOV and margin. Cheap traffic can raise visitors but lower conversion. Understanding the levers helps you choose the right move.
The five growth levers
| Lever | Questions | Typical tactics |
|---|---|---|
| Traffic | Are enough of the right people visiting? | SEO, paid ads, social, creators, marketplaces, email |
| Conversion rate | Do visitors buy? | Product pages, trust, checkout, payment options, speed |
| AOV | How much do they spend per order? | Bundles, cross-sells, free-shipping thresholds, premium ranges |
| Purchase frequency | Do customers come back? | Email/SMS flows, loyalty, subscriptions, new launches |
| Margin | Is each order profitable? | Pricing, sourcing, shipping costs, returns, discount discipline |
Diagnosing the limiting lever
Look at your numbers against your own history and the reality of your category:
Symptom Likely limiting lever
Lots of traffic, few orders Conversion (product pages, trust, checkout)
Good conversion, low revenue Traffic or AOV
Strong first orders, few repeat orders Retention / frequency
Growing revenue, shrinking bank balance Margin / CAC / unit economics
Growing CAC every month Traffic quality, creative fatigue, over-reliance on one channelChannels in the modern e-commerce mix
- Owned: your store, email list, SMS list, messaging channels (such as WhatsApp Business), app, community.
- Earned: SEO, reviews, press, word of mouth, organic social, creator content you did not pay for.
- Paid: search and shopping ads, social ads, creator partnerships, affiliates, marketplace ads.
- Marketplaces: third-party platforms where you sell alongside others.
Healthy stores build owned channels over time so that growth does not depend entirely on rising ad costs.
Regional realities
E-commerce looks different across markets. Cash on delivery remains important in parts of South Asia and the Middle East, while cards and digital wallets dominate in the UK and US. Buy-now-pay-later services are popular in several markets. Delivery expectations, returns culture, peak seasons (Ramadan and Eid, White Friday or Black Friday, 11.11, Diwali, Boxing Day) and preferred marketplaces vary. Always build your model on your market's reality.
Worked example: finding the lever
An online modest-fashion store in Karachi has illustrative monthly numbers:
Visitors: 60,000 Conversion: 0.8% Orders: 480 AOV: PKR 6,500 Revenue: PKR 3.12m
Repeat customer share of orders: 12%
COD refusal rate: significantComparing with its own past and similar stores, conversion and retention look weak, and cash-on-delivery refusals erode realised revenue. Priorities: improve product pages and trust (size guides, real photos, reviews), add order confirmation messages to reduce refusals, and build post-purchase flows to increase repeat orders — before spending more on traffic.
Hands-on: a growth-lever model in Python
Write your growth equation down as code so you can ask "what if?" questions quickly (all numbers illustrative):
from dataclasses import dataclass
@dataclass
class Month:
sessions: int
conversion_rate: float # orders / sessions
aov: float # average order value (net of discounts, excl. tax)
gross_margin: float # after product cost
variable_costs: float # per order: shipping, payment fees, packaging, returns allowance
marketing: float
def orders(self): return self.sessions * self.conversion_rate
def revenue(self): return self.orders() * self.aov
def contribution(self):
return self.revenue() * self.gross_margin - self.orders() * self.variable_costs - self.marketing
base = Month(sessions=60_000, conversion_rate=0.018, aov=48, gross_margin=0.60, variable_costs=9.0, marketing=14_000)
scenarios = {
"+10% sessions (more ads, same efficiency)": Month(66_000, 0.018, 48, 0.60, 9.0, 15_400),
"+10% conversion rate": Month(60_000, 0.0198, 48, 0.60, 9.0, 14_000),
"+10% AOV (bundles)": Month(60_000, 0.018, 52.8, 0.60, 9.0, 14_000),
}
print(f"Base contribution: {base.contribution():,.0f}")
for name, m in scenarios.items():
print(f"{name:45s} revenue {m.revenue():>9,.0f} contribution {m.contribution():>9,.0f}")Running it shows why levers are not equal: buying 10% more sessions also buys 10% more marketing cost, while a 10% AOV lift from bundles adds revenue with no extra orders to ship. Plug in your own numbers from your store platform and analytics.
The 2026 channel mix: what has changed
- Search is fragmenting: shoppers ask AI assistants and AI search features for recommendations, alongside Google Shopping, marketplaces and TikTok search. Clean product data (feeds, structured data, reviews) feeds all of them.
- Retail media (ads sold by retailers and marketplaces such as Amazon, noon and Daraz, and by supermarket groups) is a major budget line for brands selling through retailers.
- Automated campaigns (Google Performance Max, Meta Advantage+ sales campaigns) depend on your product feed, first-party conversion data and creative variety.
- Social commerce keeps growing where it is available — TikTok Shop, for example, operates in a specific list of countries (see the social commerce lesson), not everywhere.
Common mistakes
- Scaling ad spend before fixing conversion and margin.
- Measuring revenue growth without profit.
- Treating all traffic as equal.
- Ignoring repeat purchase behaviour.
- Copying tactics from a different market without adapting them.
A monthly growth review
Hold a short monthly review built around the levers. For each one, note last month's figure, the change versus the previous month and the same month last year, and the single biggest driver of the change. Then choose one lever to focus on next month and one experiment to run. This keeps the team from chasing every idea at once and makes progress visible. Over a year, twelve focused improvements compound into a very different business.
Questions to ask before spending more on ads
- Is our conversion rate stable or improving for each main traffic source?
- Do we know our contribution margin per order after all variable costs?
- Are first-time buyers coming back at a healthy rate?
- Can operations (stock, delivery, customer service) handle more orders?
If any answer is no, fix that first.
Growth model checklist
Key takeaways
- Revenue = visitors × conversion rate × AOV; lifetime value adds frequency and lifespan.
- Improve one lever without damaging others — discounts and cheap traffic often do.
- Diagnose the limiting lever before spending more on traffic.
- Build owned channels and adapt to your market's payments, delivery and seasons.
Check your understanding
Quick questions to lock in the lesson. They don’t count towards your certificate.
Put it into practice
Record your store's (or a sample store's) monthly visitors, conversion rate, AOV, repeat rate and contribution margin, and identify the limiting lever with evidence.
Enrol for free to save your progress
Reading is always free. Enrol to keep your place, take the final assessment and earn a verifiable certificate.