What are some good practices for creating manual test steps with large expected results?

Scott Pantall

We are just getting started using Xray for testing and we are starting out with manual tests. We have some tests that import data from a file and these tests have many Expected Results.

What are some good ways to make things easier on manual testers? Should we just have 1 Test Step with a large Expected Results section or is it acceptable to have a different Test Step for each expected result with only Step1 having an actual Action and the subsequent Test Steps having an Action like "check for this..."?

Sergio Freire - Xblend

Hi @Scott Pantall ,

let me share with you an article that provides some guidance on writing scripted tests: https://confluence.xpand-it.com/display/XRAY/Best+practices+on+writing+great+Test+cases

 

Let me give some additional feedback though; I would recommend that:

• you don't have "one step" scripted test cases.. from my experience, it normally is a sign that whoever wrote it has many assumptions or dependencies that were not explicit
• I wouldn't recommend either to have fully extensive detailed steps with a huge number of them; normally what happens here is that people will start validating everything (i.e. a bunch of requirements) in one test case and also removing the freedom for exploration. It's important that people know what they should do, but we shouldnt tell them to move 100pixels to the right because that would remove the minimum level of exploratory testing that i think we should have, even though we are following a scripted test case approach
• concerning expected results, I dont have the full details of your case but I think it's important to not have an extensive set of things to validate.. that would be probably way better addressed through test automation, as we as humans tend to get overloaded with much information (and also, it's not where we can provide added value IMO)

Regards

Sergio