I've worked with a handful of excellent QA. In my opinion - the best QA is basically a product manager lite. They understand the user, and they act from the perspective of the user when evaluating new features. Not the "plan" for the feature. The actual implementation provided by development.
This means they clarify edge cases, call out spots that are confusing or tedious for a user, and understand & test how features interact. They help take a first draft of a feature to a much higher level of polish than most devs/pms actually think through, and avoid all sorts of long term problems with shipping features that don't play nicely.
I think it's a huge mistake to ask QA to do automation tests - Planning for them? Sure. Implementation? No. That's a dev's job, you should assign someone with that skillset (and pay them accordingly).
QA is there to drive quality up for your users, the value comes from the opinions they make after using what the devs provide (often repeatedly, like a user) - not from automating that process.
Right - the best QA people need only be as technical as your user base. Owning QA environment, doing deploys, automated testing, etc are all the sort of things that can live with a developer.
They are there to protect dev teams from implementing misunderstandings of tickets. In a way a good Product Manager should wear a QA hat themselves, but I've seen fewer good PMs than good QAs....
Yup - I want to echo that the best PM I worked for also did quite a bit of QA work himself.
A deep understanding of the direct user experience of working with your product is a really valuable thing to have if you want to make users actually like the product. There's a WORLD of difference between the user experience that is mocked in a tool like figma or written down in a system like jira, and the actual live result.
As an aside, some of the most impactful engineers I've worked with also manually interact with the product incredibly often during development. If you automate a test too early with e2e tooling, you miss out on the wisdom gained by having to click through a feature 100+ times during development. Personally doing it exposes all sorts of rough edges and pain spots that your users are going to feel. Automating it makes you numb to them instead. It's a difficult balance.
Reminds me of how often I've felt a little envious as a dev of how much influence QA people had on effective specification. Whenever a spec appears a little ambiguous (or contradictory) the QA person becomes judge and their decisions effectively become law.
yes - devs are great at coding so get them to write the tests and then I, a good tester, (not to be confused with QA) can work with them on what are good tests to write. With this in place I can confidently test to find the edge cases, usability issues etc
And when I find them we can analyze how the issue could have been caught sooner
I've worked with a handful of excellent QA. In my opinion - the best QA is basically a product manager lite. They understand the user, and they act from the perspective of the user when evaluating new features. Not the "plan" for the feature. The actual implementation provided by development.
This means they clarify edge cases, call out spots that are confusing or tedious for a user, and understand & test how features interact. They help take a first draft of a feature to a much higher level of polish than most devs/pms actually think through, and avoid all sorts of long term problems with shipping features that don't play nicely.
I think it's a huge mistake to ask QA to do automation tests - Planning for them? Sure. Implementation? No. That's a dev's job, you should assign someone with that skillset (and pay them accordingly).
QA is there to drive quality up for your users, the value comes from the opinions they make after using what the devs provide (often repeatedly, like a user) - not from automating that process.