How we work | Zapolu

A scoped engagement, agreed in writing, with the engineers who will actually do the work. Phases are predictable, every deploy can be rolled back, and you talk to us directly instead of through an account manager.

Many agency engagements are opaque: a pitch deck, a contract, silence, then a delivery. This page describes how ours run. We adjust as needed, but this is the default.

1. Entry audit (week 0, small fixed price)

The first step is a short intro call, free and about 30 minutes, where we talk through what you need and plan the cooperation. The usual first engagement after it is this audit: send us your store’s URL and we measure from the outside of your stack: Lighthouse, Core Web Vitals from field data where available, and the three fixes we’d make first. We explain the findings on a 30–60 minute call, and we ask to see your repository, your production environment, and your analytics. If we are not the right team for the job, we say so on the call.

You walk away with: the measured numbers, the top-three fix list, and a one-page scope sketch in writing that says what we’d do, what it would cost, and how long it would take.

2. Deep-dive audit (weeks 1–2, fixed price)

When the entry audit points at something bigger, this is the follow-up. We read your code, your infrastructure, and your analytics: real-user metrics, server-side traces, a bundle and asset audit.

You walk away with:

A written report (PDF or repository issues, your choice).
A prioritized fix plan, with each item tagged by effort estimate, expected impact in milliseconds (and in conversion, where the data supports an estimate), and risk.
A go/no-go recommendation if the audit was scoping a bigger project.

3. Plan (half a week, included)

We agree on what we’re doing, in what order, and with what success criteria, and that agreement becomes the engagement document. Any disagreement gets settled in writing before the build starts, not at week 8.

4. Build (weeks 2–N, weekly invoiced)

Implementation happens in your repo, on your branch model, in small reviewable PRs, so you see every change before it merges. A weekly written check-in covers what shipped, what’s next, and what’s blocking.

Your time commitment: one technical point of contact, 1–2 hours a week to review and decide. Less works if you prefer to delegate; more if you want closer involvement.

5. Cutover (final week, included for migration-shaped projects)

Launch is the riskiest part of any migration, so we parallel-run with shadow traffic rather than flipping everything at once. A written go/no-go checklist is signed off before cutover, a rollback path stays usable for 30 days, and we are on the call during the cutover itself.

6. Operate (first 30 days included, ongoing optional)

Runbooks, monitoring, alerting that only fires when human intervention is needed, and a written post-mortem template. After 30 days you can take it over internally or keep us on a small retainer for incident backup and ongoing tuning.

What you get from us, every time

Direct access to the engineers writing the code. The person git blame points at is the person on Slack, not an account manager relaying messages.
A paper trail you can act on: scope, plan, reports, and post-mortems all land in writing.
Deploys you can undo. If something goes wrong in the middle of the night, somebody can answer “what happened” and roll back. That someone may have to be you eventually; we make sure it can be.

What we ask of you

One technical point of contact who can make decisions directly.
The access we need to do the work: repository, read-only production, third-party dashboards. Usually less than you’d expect.
Honest information about what’s broken, what you’ve already tried, and what the organizational constraints are. The project goes faster when we’re not discovering these things mid-build.

Pricing

Entry audit: small fixed price, agreed upfront (step 1 above).
Deep-dive audits: fixed price, 5–10 business days, scope-dependent.
Build engagements: weekly invoice, typically 4–16 week commitments.
Retainers: monthly, fixed hours per week, three-month minimum.

We avoid open-ended time-and-materials billing: we quote fixed prices where possible and weekly-capped rates where not.

The working agreement behind this page is public and CC-licensed. The counterpart to this page: What we don’t do.