Working together
No account managers, no handoffs between a backend team and a frontend team. The person you scope the system with is the person who writes it and answers the pager.
- 1
Scope
One call, then two days
Thirty minutes on the data sources, the channels and where money is made or lost today. Then I read your reports and write it up.
You getA written plan: the data model, the risks, the order to build in, and the number each step should move.
- 2
Build
Weeks, not quarters
One engineer writes the backend, the frontend and the deployment, and ships to a real environment every day. Migrations run against live data, never a rewrite from scratch.
You getThe first measurable change in production inside the first two weeks, with a before and after.
- 3
Operate
Monthly retainer
Watchdogs on every source and channel, a runbook, and the same person answering when something looks wrong.
You getA monthly report of what each change earned, and a system that is still correct a year later.
Three ways to engage
Every engagement starts with the same call. What follows depends on where your system is today.
Build a system
A catalog, a pricing engine, a marketplace integration or an enrichment pipeline that does not exist yet. Fixed scope, daily deliveries.
Fits: Stores adding channels faster than the team can keep up.
Run what exists
A retainer for a system already in production: monitoring, fixes, marketplace API changes, and the next improvement each month.
Fits: Operators who want one accountable engineer, not a ticket queue.
Rescue a half-built one
A system someone else started, with a deadline attached. Audit first, then stabilise, then finish the parts that earn money.
Fits: Teams whose previous build stalled or whose developer left.
The discovery call is the first deliverable.
It is not a sales pitch. By the end of it you have concrete ideas: where margin is leaking, which manual task costs the most per month, what to automate first and what to leave alone. That is yours to keep, whoever builds it.
Some clients take that plan to a cheaper team. Some of them come back a few months later with a system that half works and a deadline. Both are fine. The second case is where the most interesting rescues come from.