1

We work for a bank, and our team is responsible for doing the UAT testing of a product or feature in our mobile or web-banking systems. Usually, the IT handles the development and then it is released to QA (part of IT) to conduct tests. Now these tests are not visible to us a UAT team because we write our own tests.

The problem is that most of the time the QA deliverable is not to the standard expected for the UAT and we end up finding issues that should have been found during QA, resulting in us spending more time and effort.

What would you recommend as a handover process? What are the deliverables that we should expect from QA? A list of test cases? A list of findings? Pass/fail scenarios? What would you recommend?

1
  • 3
    What are the reasons for this separation? Why don't you work together with the first group with the single goal of perform testing ("the process of evaluating a product by learning about it through experiencing, exploring, and experimenting, which includes to some degree: questioning, study, modeling, observation, inference, etc.")? – João Farias Sep 15 '20 at 15:02
2

Obviously, we can't solve this remotely, we'd need to actually sit and talk together with their team and think together why the testing is not what it should be (whatever it means in your context). That's the whole point, communication, talking about what they do, what they don't do and why and what needs to be done in order to deliver a better product. It's a team effort, not an effort of one person, one manager or even one team in this context. You need to bring people together and approach it as a problem you all have, not just them vs. us, that creates a blaming culture, but will hardly help you deliver a better product in the long term.

What would you recommend as a handover process?

Nothing. More formal rules won't help much in the long term. Working together, pairing, more communication will help. I know, it's more difficult, definitely harder than throwing more rules on them and ordering them to follow them.

What are the deliverables that we should expect from QA? A list of test cases? A list of findings? Pass/fail scenarios?

Testing is a performance, not one thing, a state, a document etc. you can hand over. Just think about it: you require them to give you a list of test cases. You'll definitely get some, especially if they're affraid of losing their job. But it says nothing about whether or not the test cases are appropriate in the context. Therefore, it will not help you with the overall quality.

I think what you're describing is a typical problem of teams that do now work in an agile style, but rather have their own departments and strictly defined roles. This creates these silos where people do not talk to each other much, blame each other, don't even know what other people do etc. As a (partial) remedy, I'd recommend:

  • sit down with their QA team, try to understand what testing they do, what they don't do, why, what they are missing if something (maybe you don't provide them with enough business context, so they don't know what the important things for business people are)
  • try to talk to each other more, perhaps you can bring them over to your meetings or standups (if they exist), you can plan testing together, create test data together
  • you might pair with them when testing, this might be the most straighforward way of finding what tests they run and how they go about designing them and executing them

I even feel this doesn't need to be a slow or lengthy change. If people are willing to work together, you might resolve these problems fairly soon. Sustaining (some of) these changes will be harder.

2
0

First part of the problem is QA test cases are definately missing what is your team is currently verifying in UAT.. There are multiple reasons for this.

  1. QA who is testing does not have end of end understand. They only think their testing from their component testing point of view.

  2. QA might have less understanding about functionality in test.

  3. There is no or less test case review is done from either business or product owners.

  4. QA env is not setup for end to end test.

  5. UAT team is not involved in QA activities.

What you can do fix it..

  1. Involve UAT team in QA cycle.
  2. Review meeting of test cases should be held with UAT teams as well to inclose missing cases.
  3. Any end to end test infra hurdle should be removed.
  4. Regression test should be automated and should be updated regularly and should be presented as exit criteria.

Hope this helps..

2
  • 2
    If the tests itself are missing, then how would automating regression test help in fixing the problem? – JAINAM Sep 16 '20 at 4:09
  • Based on Scenario we can have our own set of test cases and even the scenarios are missing we can use the requirments to list down the scenario and hence the test cases – Hemant Varhekar Oct 27 '20 at 9:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.