Getting ready to ship: checklist, monitoring, gradual rollout and a rollback plan

What to check before going to production: quality, security, performance, monitoring, gradual rollout, rolling back, and the first hours after release.

Methodologymetodologiarilascioproduzioneingegneria del software19 November 2026 · updated 19 November 2026 · 2 min read

Releasing to production is the moment when everything you took for granted is tested by real users. Preparing for it does not mean adding ceremony: it means having answered, beforehand, a few questions that are expensive to answer afterwards.

The checklist before you ship

Quality. Tests pass, no serious known errors, the review is done, the change has been tried in an environment similar to production.

Security. No secrets in the code, dependencies checked, access controls verified, sensitive data protected.

Performance. Response-time requirements are met under a realistic load.

Data. Migrations have been tried on a copy, with a plan for the data in case of rollback.

Documentation. Release notes, instructions for whoever runs the service, support told what is changing.

Accessibility and compliance, where required.

Monitoring before the release

Before shipping you know how you will notice that something is wrong. What you watch: error rates, response times, business indicators (completed orders, successful logins). Who receives the alert and at what threshold. A release without monitoring moves the discovery of problems onto the users.

Gradual rollout

Where possible, you avoid switching everyone at once: you expose the change to a share of users, watch, then widen. Feature flags let you separate shipping the code from turning the feature on. How fast you widen depends on the cost of a mistake.

The rollback plan

You write it first: when you decide to go back (which signals), who decides, how it is done, how long it takes, and what happens to the data written in the meantime. A rollback that has never been tried is a hope. You also note when it is not possible (an irreversible migration) and decide in advance what you will do in that case.

After the release

The first hours matter most: someone watches the signals, support knows where to look, the team can be reached. A release is not finished when the code is in production but when the signals are stable. Then you record what you learned.

Warning signs

  • Releasing without knowing how you would notice a problem.
  • No rollback plan, or one that was never tried.
  • Releasing a high-risk change to everyone at once.
  • Releasing at a moment when nobody can step in.
  • Data migrations never tried on a copy.
  • Nobody assigned to watch after the release.

An example

A team releases a new payment flow to ten percent of users. Monitoring shows completed payments dropping in that group. The rollback plan says to turn the flag off: they do it in two minutes. The cause is found calmly, and no other user ever saw the problem.

When to use it

When you are preparing to release to production, need a pre-release checklist, are setting up monitoring, or are planning a gradual rollout or a rollback strategy.

This method derives from the open-source agent-skills family by Addy Osmani (MIT licence), rewritten and adapted.

This article is an OIAR skill

What you have just read is the content of a skill from the OIAR catalogue, written for people. Knowledge like this, but richer and deeper, written for the agents, drives the interviews, the analysis and the report inside the platform. Skills are used in OIAR: to try this one on a project of yours you need an account.

shipping-and-launchA method for preparing a production release: checklist (quality, security, performance, data, documentation), monitoring, gradual rollout, rollback plan, watching after release. Use it before going to production.