For Thomas Karl, the future of lean agile testing lies in predictive quality engineering. After the reactive and the proactive approach, this next stage combines big data analytics, AI systems and on-demand test environments from the cloud.
Thomas Karl is Head of Quality Transformation Services and Thought Leadership Portfolio Lead of Software Engineering for Germany, Austria and Switzerland. He is an organizational change manager, LSS Master Black Belt, SAFe® SPC5, ICAgile Enterprise Agile and Transformation Coach, ISTQB Full Advanced Level certified, a Kanban trainer and a systemic coach. His work centers on quality management and on building lean-agile organizations, from the team level up to strategy and portfolio. He has more than ten years of experience with lean-agile transformations and process improvement in large, complex international projects. He writes, speaks at international conferences and advises boards and executives on digital transformation with agile methods.
Future Challenges for Software Testing
I find it very hard to give a general answer to this, because so much depends on the context and the company. Broadly speaking, I think software has become a fixed part of our lives, and quality has become the central requirement. In many companies, though, that importance isn’t reflected in how things are prioritized, and it isn’t yet in many people’s heads either. That has to change. In my view, the quality mindset has to spread and carry more weight.
I also think software testing is still partly looking for its place in the agile world. Going forward, software quality assurance will have to cover process quality as a matter of course.
Testing faces many new challenges, but in its modern form, as quality engineering, it also opens up new opportunities for the business side.
You can read more about the different levels of testing on the HoaQ Blog.
A Framework for Lean Agile Testing in Large Organizations
Together with Nico Liedl and with input from many other colleagues, I wrote the book “The House of agile Quality”. Drawing on years of hands-on experience, we describe a structured approach to making software quality assurance work in an agile environment, and what to watch out for during an agile transformation. The HoaQ framework is practical and proven, and it has already helped many companies deal with the challenges of large solution development. It closes the gap between agile frameworks on one side and test frameworks on the other, and it connects the strategic and tactical levels of lean quality management.
The concept has five parts.
The base is the test foundation. Without the ways of working and methods it contains, quality assurance won’t deliver the results you want.
Three pillars stand on top of it:
- Roles: Which roles will be needed, what technical knowledge and understanding people need, and how to set up and develop the organizational structures.
- Approach: Choosing the agile scaling framework and the test approach, that is, what gets tested within an iteration (in-sprint) and which parts of the software are tested afterwards (cross-sprint). This pillar also covers change management, so that the changes last.
- Tools and supporting measures: The technical and process questions around choosing tools, plus what to consider when defining test environments, managing test data and building an end-to-end tool pipeline.
The roof, Lean Agile Quality Engineering methods, completes the house and holds the more advanced methods and techniques.
Each of the five parts is split into further sub-areas. That keeps the transformation manageable and lets you develop a lean QM strategy step by step.
Predictive Quality Engineering: How We Will Test
“Predictive” is the key word here, and I believe it will change the industry. We started with a reactive test approach and work proactively today, and the trend is clearly heading toward predictive quality engineering. That means the combination of big data analytics, AI systems, robotic process automation and on-demand test environments from the cloud will take software quality assurance to a completely new level. AI-based optimization of test coverage, self-healing test automation and similar approaches are already in successful use, and I’m convinced this is only the beginning.
In my view, using these technologies consistently is also a must, because high-quality software is one of the cornerstones of digitalization. Software is part of our lives and will matter even more in the future: smartphones, fitness trackers, navigation systems, cars, ticket machines, supermarket checkouts, the computer at work and plenty of other everyday things. All of it only works as intended if the software behind it works without errors. People no longer simply forgive errors, and quality has become the central requirement for software.
More on why quality engineering matters is on the HoaQ Blog.
How Testers and Test Managers Can Prepare Today
For me, it comes down to continuous learning. I see three strands that matter a lot:
- Knowledge of the different software development models and agile scaling frameworks (Scrum, SAFe®, etc.).
- Knowledge of technologies, both to understand the applications under test and to understand test automation solutions and use them in a targeted way.
- Process knowledge and an understanding of the business workflows, which matter more and more in times of digital connectivity and end-to-end digitalization.
Beyond that, I think looking beyond your own field on a regular basis is an important success factor for quality engineers.
Frequently Asked Questions
Why doesn’t software testing often have a firm place in agile organizations yet?
Because the importance of software quality isn’t reflected in the priorities of many companies. Software has become an integral part of everyday life, making quality a central requirement, but the quality mindset must first gain wider acceptance and significance. In addition: Quality assurance must not be limited to the product but must, by default, also take process quality into account.
What changes when testing becomes quality engineering?
The focus broadens from finding bugs in the product to the quality of the processes, thereby opening up new possibilities for the business side as well. Errors are no longer tolerated, whether in a smartphone, a ticket machine, or a supermarket checkout. High-quality software is therefore considered a crucial cornerstone of digitalization, not a downstream testing task.
What gap are frameworks for agile quality assurance meant to bridge?
They sit between two worlds: agile scaling frameworks on one side and pure testing frameworks on the other. The HoaQ framework described in 2022 filled precisely this gap and additionally linked the strategic and tactical levels of Lean Quality Management. This primarily addressed the challenges of large-scale development projects (Large Solution).
How is a structured approach to quality assurance in an agile environment organized?
The HoaQ concept uses the image of a house with five areas. The base is the testing foundation, comprising working practices and methodological approaches. Resting on this are three pillars: roles and organizational structures; an approach incorporating a scaling framework, testing procedures, and change management; and tools and supporting measures. The roof is formed by more complex Lean Agile Quality Engineering methods.
Does a quality strategy in a large program have to be implemented all at once?
No. The five components of the house are broken down into further sub-areas so that the transformation remains manageable. The Lean QM strategy thus evolves iteratively, step by step, rather than as a one-time overhaul. The “Approach” pillar explicitly includes change management to ensure that the changes are sustainably embedded and do not revert after a short time.
What is meant by Predictive Quality Engineering?
It describes the third stage following the reactive and proactive test approaches: quality is predicted rather than merely tested. This is supported by a combination of big data analytics, AI systems, robotic process automation, and test environments that are created on demand from the cloud. AI-based optimization of test coverage and self-healing test automation were already in use by 2022.
What knowledge do testers and test managers need to keep up with this development?
Three areas are key: knowledge of software development and agile scaling models such as Scrum or SAFe; technical knowledge to understand both the applications being tested and the automation solutions; and process knowledge, including a technical understanding of business processes. The latter is becoming increasingly important with end-to-end digitalization. Continuous learning and thinking outside the box are key.
