An ecommerce marketing strategy is only as good as the numbers under it. Most strategies start with channels. As a strategy partner, SSO starts with your economics: what a customer is worth after every real cost, what one costs to buy, and what the business needs to cover itself and pay you.
This page describes the investigation we run in the Validation Sprint: what it establishes, how it differs for a new offer and an existing store, what you receive, what we need from you, and how the evidence becomes a sequence of work. If you would rather build a plan yourself, our guide to building an ecommerce marketing plan walks through it step by step.
What the investigation needs to establish
Two different numbers sit at the centre of the investigation. Most plans blur them, and that is where a lot of wasted spend begins.
Required economics: what the business has to earn
Your costs covered, what you pay yourself, and a surplus you have actually named. We put your pay in the numbers, because a plan that leaves it out quietly relies on a volunteer. This number comes from your cost base, not from a market benchmark.
Achievable economics: what the offer can support
What your offer, customers and costs can realistically produce. That means your average order, the margin left after cost of goods, shipping and returns, what traffic costs in your category, and how often customers come back. This number has to be evidenced, not assumed.
The gap between the two is the diagnosis. If what is achievable clears what is required, the question becomes how fast and in what order. If it does not, more marketing only makes the gap more expensive. Our article on why your ROAS target is a requirement, not a price explains the same idea from the advertising side.
Required versus achievable, on one store
Hypothetical figures for illustration. A store needs $6,000 a month of gross profit after media to cover its fixed costs and pay the founder. Its average order is $80, and after product, shipping and payment fees each order contributes $48 before media.
If a new customer can be bought for $30, each first order leaves $18. The store would need about 334 first orders a month to reach its target from new customers alone, before counting any repeat purchase. That may or may not be realistic, and testing is how you find out.
If a new customer costs $60 to buy, every first order loses $12. The target is out of reach at any volume unless the offer, the margin or the repeat rate changes. That is a strategy finding, not an advertising one.
We then check three things in the order the money flows. Can the business afford a customer? Does the site convert the traffic you already pay for? Can attention be bought at a price that works? If one does not hold, we stop there and tell you. What to fix first, in the order the money flows sets this out in more detail.
A real case shows why the order matters. In January 2026, a customer of Apothecary by Justin was worth about $55 in gross profit and cost about $400 to buy. Required and achievable were nowhere near each other, and three agencies had already tried to optimise the ads. The fix was the offer, which took his average order from $69 to $137 by August 2026.
Testing a new offer versus diagnosing existing trading
The Sprint runs differently depending on how much evidence already exists. Sparse sales data is not a reason to turn you away. It changes how the four weeks are spent.
| New offer or little sales data | Existing trading | |
|---|---|---|
| The main question | Will people buy this offer at a cost the business can carry? | Where do the current economics break, and what is the first fix worth? |
| Where the evidence comes from | Campaigns we run during the Sprint to buy data, with full tracking | Your existing store, advertising and sales data, plus one targeted test |
| Media | Test media is additional, agreed and capped before anything goes live | The test is scoped with you before it runs |
| What low volume means | The test exists to create evidence where there was none | Small numbers widen the uncertainty, and we say how much weight a result can bear |
When there is little or no data
In week one we go through your numbers, not your ads: cost of goods, shipping, returns and what you pay yourself. In week two, campaigns go live to buy data, with real money and full tracking. That week is not meant to be profitable, and we tell you so before it starts.
In week three we test against what week two showed: angles, audiences and offers. We follow what moves and stop what does not. In week four you get the answer.
When the store already trades
We audit what is there first: the offer, the margins, the site and the advertising, using your real numbers rather than platform summaries. Then we run one targeted test on the question that matters most, instead of a spread of small changes that cannot be read apart.
Either way, four weeks of data is a starting point, not a verdict on the next five years. We say how confident each finding is and what further evidence a later commitment would need.
What the Validation Sprint gives you
Whatever the verdict, the work is yours. It is not licensed to us. Use it with us, with another provider or on your own, whether you do the work yourself, brief freelancers or hire in-house.
-
Your goals with numbers attached.
What the business needs to cover, what it needs to pay you and the surplus you want it to produce. Then what your offer, customers and costs can realistically support.
-
What is working, what is broken and what each broken thing costs you.
Put in dollars, so you can see which problem matters most.
-
Where you sit against the five things we score every store on.
A consistent view of the business, not only the part that one provider happens to sell.
-
The first ninety days, in order, with the reasoning attached.
Including responsibilities, review points and any further evidence needed before a later commitment.
-
Actual people and hours against the work.
Including your own input, so the plan does not quietly rely on a second you.
-
Every number behind every recommendation.
The workings, not a summary.
-
The recommendation itself, written down.
One of three verdicts, with the reasons.
We walk through it with you in a ninety-minute session: every number, where it came from, and the recommendation in writing.
The three possible verdicts
-
Go ahead.
We make you an offer to do the work, on one of two routes.
-
Not yet.
A fix list, in order. Do it yourself or with anyone, and come back when it is done.
-
Not with us.
No offer. You keep the plan, and we tell you exactly why.
Proceeding, testing further, changing direction or stopping can all be the right call. Roughly six in ten completed Sprints are not followed by an offer from us (September 2026 record, not all enquiries and not a promised future rate). Those founders still leave with the plan in their hands.
From evidence to a costed sequence of work
A finding becomes a plan when every step has an order, a cost and an owner. Where one fix only pays off after another has landed, we say so rather than let you find out. Where the evidence is not there yet, we say what the next test has to prove before you spend.
- 1
Order the fixes by the money
The first step is the one that earns the first dollar, not the one that is easiest to sell. For Apothecary by Justin that meant the offer first, then the ads.
- 2
Cost each step
Media, delivery, production and any website rebuild, with what each step depends on and what it is expected to be worth. Our guide to setting an ecommerce marketing budget covers the same categories for founders planning on their own.
- 3
Put people and hours against it
Who does each piece, including you. The plan should work whether you hand it to us or to someone else.
- 4
Set review points
When the numbers get checked, and what would make you continue, change direction or stop.
What we need from you during the Sprint
- Access to your store and advertising accounts. The four weeks start from the day we have it.
- The real numbers: cost of goods, shipping, returns and what you pay yourself, including the ones you are not proud of.
- Funding for capped test media, if the business is new and we need to buy data.
- Time to go through the numbers with us at least once, and to attend the ninety-minute readout.
- A willingness to hear things you may not enjoy in week four.
Our approval is a separate decision
A positive finding does not automatically become an offer. Before we propose ongoing work, we assess three things separately: whether the opportunity is credible, whether you can fund and carry out your side of the plan, and whether we can deliver it at economics and capacity that work for us.
Turnover and business type do not decide the route or the fee. A profitable small store and an established store with broken economics can need very different things, so the route follows the evidence.
If we do make an offer, it is on one of two routes. Launch builds the economics: twelve months, 30% of agreed eligible product gross profit against a $1,500 plus GST monthly minimum, and you pay the greater of the two, never both. Accelerate expands economics that already work: a twelve-month initial agreement, invoiced monthly in advance, scoped to the work and agreed in writing.
For selected Launch businesses we fund an agreed part of the early delivery work, written down before you commit. There is no equity involved, and it does not cover your media, stock or operations. If you go on together, the $4000 Sprint fee is credited across invoices 11 and 12 of the initial twelve months, $2,000 each, without reducing the first ten.
For how fees compare across providers, see what an ecommerce marketing agency costs. If you are weighing up providers for the first time, hiring your first ecommerce marketing agency covers what to ask. For a wider view of everything SSO does, see our overview for Australian product businesses.