Your Automation Strategy needs to change alongside the product lifecycle

I feel Automation can add lots of value if used in the right way. Before you get anywhere near automating — answer the question…

What you’re checking and WHY?

Sometimes I see a lot of talk about automation, but less so about whether what we’re checking is valuable.

Feature Tests > Build Mode

Initially you might have a lot of automated checks when building out the feature as lots of things are breaking all over the place and you need quick feedback to fix things. These are what I’d call feature tests.

Build > Maintenance Mode

• Later on you may move from building to maintenance…ensuring old features don’t break with new changes. This is where you can tag your tests as smoke/regression/feature which then means you can run these with the correlated activity e.g. regression. Overall reducing your run times, lowering the costs/time/effort without compromising on quality.

Maintenance > Deprecate

Then when the feature is no longer used….why run the automation at all? Let’s just delete it.

Automation strategy needs to be thought about alongside the product lifecycle (e.g. build/maintain/deprecate). It’s a constant review process and change in strategy