QA Automation Playbooks

A playbook is a procedure. It starts with what you need before you begin, then walks through numbered steps in the order you would actually do them, and finishes with a worked example, the mistakes teams make most often and the practices that prevent them. You should be able to keep a playbook open beside your work and follow it from start to finish.

A guide does a different job. Guides explain a topic — what end-to-end testing is, how manual and automated testing compare, how to choose a tool — so you can make a decision. Playbooks assume the decision is made and show you how to carry it out. Most teams read a guide first and come back to the matching playbook when it is time to do the work.

Every playbook uses the same e-commerce product as its worked example, so the steps are easy to compare from one playbook to the next. Each ends with related guides, key figures and sources for further reading.

To use a playbook with your team, read it once end to end before you start, then agree who owns each step. Work through the steps against one real feature or release rather than the whole product, and keep notes on where your situation differs from the worked example. Those notes become your team’s own version of the procedure, which you can reuse for the next release.

The three playbooks also fit together. A strategy decides what to automate and in what order; good test cases are what you automate; and a regression cycle is how those tests protect each release. If you are starting from nothing, work through them in that order. If you already have a suite, start with whichever step is causing the most rework today.