About
Better decisions, before the money moves.
SellZyme is being built to help marketplace teams make better advertising decisions before money is spent — not simply report on mistakes afterward.
Why we're building this
Amazon advertising has no shortage of automation. What it lacks is foresight. Almost every tool on the market works the same way: it checks a condition, changes something, and waits to see what happened. The account funds the experiment, and the findings arrive after the bill.
That struck us as backwards. By the time most rules fire, the information needed to know roughly what will happen already exists — the keyword has history, the placement has a conversion rate, the product has a margin. The tools just don't use it.
So we're building the layer that does: a system that models the likely outcome of each available action, compares them, and only then acts — inside limits you set, with a record you can audit.
What we believe
Four convictions
A decision is better than a reaction
Most advertising tools are very good at telling you what already happened. That's reporting, and it arrives after the invoice. We think the useful moment is before the money moves.
Profit is the only honest metric
ACOS knows nothing about your cost of goods. A campaign can hit its target and lose money on every unit. If a system doesn't understand contribution margin, it isn't optimising your business — it's optimising a ratio.
Autonomy has to be earned
Nobody should hand an account to a black box. Every action we take is logged with the prediction that justified it, and scored against what actually happened. Trust should be evidence, not a leap.
Knowing when not to act is the skill
The strongest available action is frequently no action. A system that can't express restraint will always find something to do, and most of it will be noise.
Where we are
SellZyme is early. We're building with a small group of partners — sellers, agencies, and brands who care about the same problem and are willing to tell us when we get it wrong.
We'd rather say that plainly than dress it up. If you want a platform with a decade of case studies, we're not it yet. If you want to shape one, this is the interesting moment.
Want to shape what we build?
We're working with a small group of partners who care about this problem. Tell us about your account.