It is easy to describe what we do. Four numbers, a written plan, a base fee plus growth share, month to month. The list that tells you more is the one below, because refusals cost money and nobody publishes them unless they intend to keep them.
Every item here has come up in a real conversation with a store. Some are about how we get paid, because pricing decides which advice you get. Some are about ownership, because the accounts your business runs on should not sit with a supplier. Some are about proof, because the fastest way to sell marketing is to invent evidence, and we would rather write nothing than write that.
None of it is meant as a criticism of anyone else's model. It is a description of ours, so you can decide before you spend an hour on a call whether we are the wrong fit. Being the wrong fit early is cheap for both of us.
We will not price the work as a share of your spend or your sales
Our fee is a flat monthly amount, scoped in the plan, and it does not move when your ad spend moves or when your revenue moves.
A percentage of ad spend pays an agency more for spending more, which puts the person advising you on budget on the wrong side of the question. Every recommendation to raise budget then arrives with a financial interest attached, and every recommendation to cut a campaign costs the person making it money. You should not have to weigh advice that way.
A percentage of existing sales looks fairer and has its own problem, which is why our growth share only touches revenue above your baseline. It charges you more for revenue that was already coming, including repeat orders from customers you won years ago, and it makes the fee unpredictable in exactly the months when cash is tight.
A base fee plus growth share has one honest weakness worth naming: if the work turns out to be less than we scoped, you are still paying the scoped amount. That is why the scope is written down before you agree to anything and why the arrangement is month to month. If the work is done, you should be able to stop, and you can.
Ad spend goes to the ad platform, from your account. We never take a cut of it.
If an arrangement makes it hard to leave, that difficulty is the product.
We will not put your accounts in our name
Your store, your ad accounts, your Merchant Center, your analytics, your email platform and your domain stay in your name. We work inside them with access you grant and can revoke.
This matters most on the day someone leaves. If an agency owns the ad account, the conversion history, the audience lists and the learning that budget paid for stay with them. Starting again elsewhere means starting the data from zero, which is a real cost, and it becomes a reason you cannot leave.
It also matters while the work is running. Accounts in your name mean you can see what is happening without asking, check the spend yourself, and give a second opinion access to the same screens we look at.
The same applies to what we build. The plan, the campaign structures, the feed rules, the flows and the page changes are yours, and they stay in place if you stop working with us. We would rather keep the work by moving your numbers than by holding something you cannot take with you.
If an arrangement makes it hard to leave, that difficulty is the product.
We will not promise a number we do not control
We will not commit to a revenue figure, a return on ad spend, a ranking position or a delivery date for a search result. Not because we lack confidence, but because those outcomes depend on demand, competition, pricing, stock, seasonality and platform decisions that no supplier controls.
What we will commit to is the work: what gets built, in what order, and what we expect it to move. The plan says which lever each fix belongs to and what we would look at to judge it. Then the monthly report shows what actually happened, including when it did not work.
The related refusal: we will not quietly drop a failed test out of the reporting. If we recommended something and it did not do what we thought, it appears in the report with that description.
We will not sell you a channel before we know where the leak is
If you arrive asking for more traffic, we may still tell you that traffic is not your problem this quarter, and we will say so in the plan before you have paid us anything.
The reason is arithmetic. Revenue is visitors multiplied by conversion rate, average order and repeat purchases. Buying more visitors for a store that loses them at a known step means paying twice for the same leak, and it makes the underlying problem harder to see because the totals go up.
That is not a rule against traffic work. Plenty of stores genuinely need more visitors, and when that is the case the plan says so and we build it. It is a rule against selling the channel we happen to be good at to a store whose problem sits somewhere else.
The same applies in reverse. If your product pages and checkout are already doing their job, we will not invent conversion work to fill a scope. The plan will say the pages are fine, which is a legitimate and frequently useful thing to be told.
We will not manufacture proof
No invented statistics, no borrowed benchmarks presented as our own findings, no case studies about stores that do not exist, no logos on a page for companies we did not work with, no reviews we wrote ourselves, no bought followers, no fake urgency counters on your product pages.
That last one deserves emphasis because it gets requested. Stock counters that are not connected to stock, timers that reset when the page reloads, and review counts that include reviews from other products all work in the short term and they all corrode the thing that makes a store worth returning to.
On our own pages, the same rule applies to numbers. Where a number is needed, it is either your number or a clearly labelled worked example with made-up figures used to show the shape of a calculation. If we cannot stand behind a statistic, it does not appear.
This costs us. Made-up numbers are persuasive and everybody knows the format. It is still the easiest rule we keep, because a claim we cannot support is a liability the first time somebody checks it.
We will not take on a store that is not selling yet, or one we cannot help
We work with stores that already sell online. The reason is practical: our method starts by finding which of four numbers is weakest, and a store with no orders has no numbers to read. Everything we would offer such a business would be a guess dressed as a diagnosis.
We also turn down work where the constraint is not marketing. If the margin cannot support any customer acquisition, if the product is not available to ship reliably, or if the category is one we do not understand well enough to be useful in, the answer is no rather than a scope written to look busy.
And we will not take on a store where the person we would work with has no authority to change anything. Fixes that sit in a queue for approval for six weeks are not fixes.
When we say no, we try to say what we would do in your position, because that costs us nothing and it is the point of writing a plan in the first place. The free plan is yours whether or not we work together, and we would rather it be genuinely useful than a hook.
What we do instead
The positive version of the same list is short. We write the plan first, free, so you can read the reasoning before any money changes hands. We quote a base fee plus growth share for the scope in that plan. We work month to month, so the reason to keep going is that the numbers moved.
We work inside accounts you own, we report in the same four numbers every month on one page, and we tell you what did not work alongside what did. When a fix needs a developer, we say so instead of pretending it is a settings change. When a lever is already fine, we say that too and put the scope somewhere useful.
If that reads like the arrangement you want, the plan is the place to start. If it reads like the wrong fit, this page has done its job.