Agile testing

Stop Wasting the Efficiency Potential of Test Automation and DevOps

Guest Post from Tasktop, by author Wesley Coelho, Sr. Director of Business Development.

Everyone knows the dual productivity powers of automated testing and DevOps practices are essential for driving large-scale software delivery in high performing teams. Both are about efficiency, quality and optimizing the software value stream. Yet despite the existence of these proven methods, many organizations are still struggling to master enterprise-scale software delivery.

Test automation helps testing teams kiss goodbye to repetitive manual testing tasks to focus on tests that run quickly and can be kicked off repeatedly by a build process. DevOps bridges the disconnect between the practitioners writing the code and those deploying it, and brings organizations closer to the customer to deliver better software faster. Together they have improved software delivery process immeasurably. Well, kind of.

The daily lives of developers, operations and testers have been improved, but often only to a point. In other words, the huge ‘efficiency potential’ of DevOps and test automation isn’t being fully realized. And the reason for that is the software value stream is broken. The best-of-breed tools required for governance of the planning, testing, and operation of deployed software do not naturally integrate.

Why does that matter? Because the fundamental foundation of creating software is about multiple teams and disciplines collaborating on projects to make sure software is high quality, what the customer wanted and in production ASAP.

Yet this cross-discipline collaboration is complicated because developers, ops,­­­­­­­­­­­­­ testers, project managers, product managers, business analysts etc. are working in their own way in their own tool with their own processes. And they’re happy that way. But what if they’re wasting time manually sending and searching for information or making decisions based on incomplete, inaccurate, or out-of-date information without realizing it? What if they’re not being as efficient as they think they are?

Take testers, for example. In principle, test automation should mean testers have more time to create more thorough tests and test more edge cases. However, in reality, the time they’ve recovered thanks to clever test automation software has now be lost to laborious manual communication with the rest of the value stream to find out what’s going on. Precious time lost to email threads, boring status meetings, eye-blurring spreadsheets, duplicate entry and so forth. Moreover, how do they know they’re even testing the right thing?

DevOps is fine though, right? Developers and ops are happily married and bound by CI/CD, churning out code and deploying at speed. Maybe not. Unless their tools are connected with the rest of value stream, they’re still forced to chase up vital information via archaic communication methods. They may as well pretend they’re in ancient Rome and use a wax tablet.

And the need to pass wax tablets around to share information becomes even worse in DevOps environments with small batch sizes and increasingly frequent releases. Furthemore, if DevOps is too far removed from the customer-facing folk, there’s a good chance they’re not even building the right thing. And that’s not just time wasted on pointless product enhancements, that’s time wasted on upsetting the customer.

All of this unnecessary waste can be solved by introducing a new layer of infrastructure – model-based integration — to connect all tools and teams in the software value stream. Just like tests need to be automated, the flow of key artifacts between different systems needs to be automated to eliminate that waste. For example, requirements to be tested need to flow to your test automation solution. And when tests fail, defects need to be created automatically within your defect management system. With a connected value stream where the right information between flows between teams in real time, you can better ensure that the right product is being articulated, defined, built, tested, delivered, and operated in production.

No longer will your initiatives be running on half a tank. You’re finally realizing the immense efficiency potential of test automation and DevOps. The perfect partnership to take your large-scale software delivery to the next level.

Guest Post from Tasktop, by author Wesley Coelho, Sr. Director of Business Development.