Decide what tests to automate.

At my workplace we are very strategic with what we automate. It’s not automating for the sake of it but a thought out strategic process. I have in the past been asked how to get into Automation and one way is to start figuring out what’s worth automating before you start with implementation.

There is this ‘What to Automate Worksheet’ that is written by Dorothy Graham based on Worksheet ideas by Angie Jones. To Go to it Click Here. This is something that we have taken to help guide us on what to automate.

In this excel it lists the features and then you assess customer risk, value of test, cost efficiency, history, which all totals into a total risk. It’s a neat way to give you a guide of what to automate first and maybe areas to skip automating at all.

A couple of things from our implementation:

• We amended it from features to business process, which is a task or activity a person would do to achieve a business goal. It is similar to a feature, yet it focuses more on the person using your product.
• If you’re automating as you go, then the history column is irrelevant, so in some cases we removed that.
• Get help from product team or those that directly built the product to help you create your business process list or features.
• We watched this Video by Angie Jones that gives an example of how to go about an automation assessment. To see this you can Click Here.

How do you approach figuring out what to automate?