Solve 3 often encountered exploratory testing challenges
So its a relatively new thought in the testing space. But one that is gaining momentum, especially because of the mindshare from testers. Exploratory testing is engaging your human intellect while utilizing your subject matter expertise.
My personal opinion is that it is a great value add, and can be used both as an early testing feedback tool, as well as the last line of defence UAT tests. However, I don't believe LARGE enterprises are ready yet to make the move to a 100% exploratory only testing. At least not yet, and certainly not in all areas of systems (Not in Systems of Record vs Maybe in Systems of Engagement).
Cut to the chase : you are lost in the jungle of business possibilities in your application, and you hit a snag. Consider the following:
1. How do you share the steps to reproduce the defect?
2.How do you add this to a progression test suite, if deemed a valuable test case?
3. In an automate first strategy, how much of the above efforts and assets can contribute to test automation?
I love the recent advancements in Tosca's exploratory testing engine, and here is a view on how the above 3 challenges are addressed.
1. Video - Tosca's exploratory testing engine, allows for video recording your session on the application. The cooler aspect is how it allows for annotations on the recording on the fly. Other than for just comments, highlights and drawing tools, the relevant parts of the recording can be tab-stopped on the timeline. Basic video trim functionality cuts out fat from the session. Plus the UI is nicely woven into a manual steps editor, so you don't have to send out a 40 mins session, where only the last 3 minutes need visual aids, and the rest can be documented as test steps.
2. Capture Interactions - Video is so old school! Want something cooler? How about we use Tosca to automatically create a narrative of the interactions? This latest addition to Tosca is exciting because of Tosca's Model based approach of test automation. For the uninitiated, Tosca has a hybrid test automation framework beneath its keyword driven interface, and hence all test logic is documented in simple business, plain linguage format. So when you use a capture interactions mode, the output is very similar to a test case, in that you get what control is being interacted with, what data is exchanged, and what input actions are being performed.
Potentially, this output can be added as a manual test case, with very little additional effort. Hence Exploratory tests can become Progression tests which over time become regression automation .... Check out the Exploratory Testing section of our Whats New Video Series page here : https://www.tricentis.com/tricentis-tosca-testsuite/what-is-new/9-2/
3. Exploratory tests are designed as manual tests, as opposed to specification based test automation. However, a snag, along with its steps to reproduce, if qualified important, may be a great addition to your progression/regression automation test suite. In automate first strategies, this manual test case resulting from the capture interactions mode of Tosca, can be automated by a simple drag and drop linking with an automation model. Again, the "model" is an interface catalog of the System under test in TOSCA.
Adopting session based exploratory testing with Tosca is easy, and contributes itself useful in both manual spec based and automated testing creation, should that be the need. In short Tosca enbables the paradigm of Session based exploration > Convert to manual test > Convert to automated test case. Interested in learning more? Join us for a webinar on exploratory testing using Tosca : https://www.tricentis.com/blog/events/webinar-catch-bugs-earlier-and-faster-with-exploratory-testing-apac/
Software Tester | Exploratory Tester | Quality Advocate | Critical Thinker | Risk Finder | Problem Solver | Mentor | Teacher | Leader | Helping Teams Build High-Quality Software
8 年Hi Arun, "the mix of TOSCA ET features as well as its Model based test automation philosophy enables adoption of ET, " A specific test tool and an automation philosophy are not components of a successful ET approach. TOSCA is probably very good at what it does but if you try to tie it to some way being fundamental to ET you'll have trouble selling your reasoning.
I help software development teams to improve their testing | Founder and consultant, Dr Lee Consulting | International conference speaker | Author of "An Exploration of Testers" (LeanPub) | Vegan for the animals
8 年The paradigm seems to add more weight to the idea that value of testing is in creating artifacts and being repeatable. The goal of testing is to provide valuable information about the product and I find session-based ET to be an excellent way to do that. It may be that a session of ET reveals to me a number of automated checks that might be valuable to create, but this is not the main goal of my ET. The session sheet is designed to be used for debriefing, to discuss what happened during the session and decide on next steps. (It also provides a credible account of the testing that was performed for the record, of course.) Why would I want to use the same thing for regression testing? Even if a session (or even a test) could truly be reproduced, why look for the same things over and over again? Having a tool that claims to be supporting ET but then offers functionality to create test cases from sessions only serves to reinforce the common view which conflates "testing" with "test cases". So it's not a question about whether this paradigm saves cycles, it's to me a paradigm that risks reinforcing bad testing ideas and will appeal to managers looking to agree to using ET but really still wanting test cases to count at the end of it.
I help software development teams to improve their testing | Founder and consultant, Dr Lee Consulting | International conference speaker | Author of "An Exploration of Testers" (LeanPub) | Vegan for the animals
8 年Not a paradigm I'm familiar with: "the paradigm of Session based exploration > Convert to manual test > Convert to automated test case"
Software Tester | Exploratory Tester | Quality Advocate | Critical Thinker | Risk Finder | Problem Solver | Mentor | Teacher | Leader | Helping Teams Build High-Quality Software
8 年Arun, I'm not sure why you would think ET is new in testing. It is over 20 years old. In trying to tie ET to a testing tool you are, unfortunately, misrepresenting the real value of ET.