Mind maps for test reporting

Hi,
I am test lead and I want to change the way we deal with big repositories of tests.
With the introduction of mind maps, the tests become simpler to view and the testers have a big picture of what is being tested.
My question is - has anyone used these in their projects before for testing?

  • If yes, how do you pull reports with these mind maps?
  • E2e scenarios - is there a way to represent them in mind maps?
3 Likes

It is possible to mark the branches with markers or icons to give a status of the tests. E.g. a green v on a branch means: all the tests in this branch passed.

My way of using mind maps for an overview of testing is to focus on the product risk. If I focus only on functionality, then I might miss other things. A useful heuristic is SFDIPOT:

  • Structure. What are the components of the system and how are they interactiing with eacht other?
  • Function. What are the features of the system?
  • Data. How does data flow through the system? This is useful for E2E or End to End test.
  • Interface. How does the system exchange data with other systems and/ or users? E.g. GUI or API.
  • Platform. Which operating system is used?
  • Operations. How will the users use the system in production and in the future?
  • Time. Which features are time dependent? E.g. does an alarm go off after 5 minutes?

It is important to keep this mind map updated on a daily basis. I frequently use Exploratory Testing. With the knowledge acquired during the tests I can adjust my approach on the spot.

More information can be found in:

3 Likes

Soooo
Since I no longer work there, I’m going to describe a well resourced project at a certain large company which has had rather little about it published.

But first Mind Maps, I think they are great for uncovering test cases , component interactions, user workflows and thus, good test-cases to go away and write up.
The Mind-Map technique is also an entire school of thought that uses graph permutation to design your test cases for you called Model Based Testing, this is a million miles away from Mind mapping. Model Based Testing is not something I know a lot about myself, mainly because I am wary of the expensive test heuristic after a bad experience using graph tooling based test design. I think MBT (Model Based Testing) has it’s place in any setting where you have either regulation as a constraint, or just have the resource available to you. Some MBT platforms use ML, another possible source of cost, but you can find a few good training or intro videos on how it works, and there are more than one or two commercial products that exist. Right here on the MOT :slight_smile: Model-Based Testing "PetClinic" example using AltWalker and Python

But, I was going to tell you about a graphical test design tool from long ago. GTP (Graphical Test Planner) was such a tool. We had an architect create a directed graph language that supported flow charting of activities and branches in a uniform language (UML) using a popular vendors UML tool we were able to draw boxed representing workflow branches and environment differences and workflow start and end criteria. Basically GTP was a cool way of exploring user interactions with our system, but in a language that removed duplication, and was a fun way to document tests without having to write stupid numbered lists of steps, because the UML would easily capture decision branches. Our GTP language was so mature that we relied on it a lot, and were eventually able to travers the graphs to determine test coverage stats. Which is a brilliant goal achieved quite well in MBT tools. We even used colors in our graphs to represent various states, and made it possible to automatically mine this data, and if you really-really wanted to automatically change element colors.

GTP never was that silver bullet, I suppose it’s still in use there at that large company, but like so many tools, the things they do teach you have value. When they stop having value, you drop the tool. I believe that Mind Maps are like that too, they are like Marmite, good, but not with every single meal.

1 Like

@han_toan_lim What has the blog post got to do with “Flat Screens” or is that a music reference that I’m just not getting.
Brilliant walk through of mind mapping techniques there. Definitely makes them less boring exercises. I like the Mind Map for things like security-problem hunts too, although, in so far as being useful, the best mind maps are the ones without a clear specific end-goal. Mind maps need to be free to capture extra info and angles, which your SFDIPOT heuristic does quite well.

1 Like

A-flat is used in music:

I used the piano, the fuga, and A-flat to describe focus.

2 Likes

LOL, I have to google every single word on that page, music is an entire language, foreign to some of us. But I understood what you wrote, about the process and ignored the musical references. I like what you did though, because a Fuga is multiple voices, IE. a collaboration . A lot of engineers are musical talented actually, so this is a pretty good way of connecting.

I use Mind Maps for the test planning aka. what are the project scope and what are the progress on that. As well as for my session reports or exploratory testing documentation if you don’t want to use Thread Based Testing or Session Based Testing. Both of which have the main purpose of being able to report on exploratory testing efforts.
So far I have separated them since there are currently no value in trying to combine the two. More work than what it saves. So the plan in updated separately from the session reports. Then I use the tool MindMup which have the progress feature where you can tag your nodes in your map with a status (like in progress / failed / passed). Then it will produce an easy overview of number of nodes in your map that has each status and it will also update parents when all their children change. Like if all of them is passed the parent will be passed to, and if any of them fail the parent will fail so you can collapse finished parts of the map. It is also integrated with Google drive which was the main driver.

1 Like