Test organization of the next decade

The test organization of the future describes a model in which traditional hierarchies and fixed roles are replaced by small, self-organized units without superiors. Testers will need so-called V-shape skills: in-depth expert knowledge combined with sound knowledge in related areas. AI tools and cloud technology are taking over routine tasks, while exploratory testing and personal responsibility are gaining in importance.

Key Takeaways

• From I-Shape to T-Shape to V-Shape: By 2030, testers will not only need basic knowledge in related fields, but also in-depth expertise in several areas in order to remain flexible.
• Test organizations of the future will work without hierarchies and fixed roles: Experts come together in small units, staff projects independently and without a traditional supervisor structure.
• ChatGPT can already create complete test plans, carry out tests and make live recommendations, but the results must be checked by humans because they are not yet consistently correct.
• Exploratory testing is gaining in importance because it covers the area that machines and AI cannot do reliably, provided it is carried out in a structured and methodical manner.
• Self-organization and personal responsibility become mandatory requirements for every tester because the hierarchy-free model no longer provides for test managers who assume responsibility on behalf of others.

Why test organizations will have to change in the coming years

The test organization of the future will be flatter, more networked and technically more sophisticated. Two drivers are forcing them to do so: AI tools such as ChatGPT and the widespread move to cloud technology. Both are taking away routine work and shifting the question of what testers are needed for in the first place.

Nicolas Nwabueze assesses the situation soberly. Even with the advent of test automation, there was concern that testers would become superfluous. This has not happened. This pattern is repeating itself with AI: the work is not disappearing, but the tasks and the necessary skills are shifting.

Those who ensure quality should therefore ask themselves less whether a machine will replace their own role. The more pertinent question is which activities can be handed over to tools and which can be given space as a result.

What ChatGPT can do in the test process today

ChatGPT can take over parts of the classic test management chain, from the test plan and test execution to the recommendation for going live. In an example we ran through, these steps could actually be mapped with the tool.

Very few people have used it in practice so far. A rough figure from the field: around 70 percent are involved with ChatGPT, but only around 5 to 10 percent actually use it. The reason is caution, not lack of interest. The results are not yet reliable enough to trust them blindly.

This results in a new ongoing task for test management: checking and validating the results of the AI. What ChatGPT delivers is a suggestion, not proof. Nobody should assume that everything is correct just because it was generated.

AI brings a clear benefit in unit testing. Developers are often reluctant to do this work, and this is where ChatGPT delivers well. If developers are relieved at this point, there is more time for features. This addresses a well-known area of tension: if everything has to be tested, the throughput of new functionality decreases.

From I-Shape to V-Shape: how the skillset is shifting

The requirement for testers is moving from deep individual knowledge to broad adaptability. Nicolas describes three stages of this development.

The difference between the T-Shape and the V-Shape lies in the depth of the secondary areas. With the T-Shape, a basic understanding of neighboring topics is sufficient. With the V-Shape, this knowledge is more substantial, so you can quickly switch to another area and take on real work there.

The reason behind this is adaptability. If a test manager with some programming knowledge suddenly finds that AI is taking over tasks from them, a narrow profile is of little help. Broader and deeper skills ensure that you continue to deliver value elsewhere.

In concrete terms, this means taking new tools and disciplines seriously. These include Azure DevOps, which few people have mastered in practice, and requirements engineering, which test management is likely to become more involved in.

Cells instead of hierarchy: an organizational model without fixed roles

The future model presented does away with roles and hierarchy. Test managers, test coordinators and testers are replaced by small units known as cells. Experts gather in these cells without superiors, and projects are staffed from within them.

A study by Forrester supports this direction with a comparison of current and future IT organizations along four axes:

• Structure: hierarchical today, small and non-hierarchical in the future.
• Skillset: more generalist today, more specialized in the future.
• Speed: slowed down by hierarchy today, much faster in future.
• Reach: today internal for own company, in future also external, i.e. as a service to the outside world.

The last point means stronger networking. An established test process can not only be used internally, but also offered to other organizations. Areas can also be provided by one service provider for several clients.

The model is more than theory. During our research, we found an IT service provider in Munich that already works in this way. Practitioners who recognized the approach also came forward in the audience and at an evening event. Cassini itself reorganized in 2019 and at least flattened the hierarchy.

Cloud technology makes organizations faster

The move to the cloud is the second major driver of change, and it is accelerating. According to a report by KPMG, 84% of companies in Germany have moved to the cloud, compared to 37% ten years ago.

The most important advantage is speed. Restoring a system from the cloud is much quicker than setting up a local computer again and again. Faster recovery means faster organization.

In addition, the cloud enables the forms of work that are now taken for granted. Remote work and new work are directly linked to this technology. Collaborative work on a document or source code works regardless of location.

Another point is the ecological footprint. Companies in the cloud tend to have lower energy consumption and lower CO2 emissions. This is another reason why the switch is attractive for many companies.

Why manual and exploratory testing is gaining in importance

Manual testing is not disappearing, it is becoming more important. Exploratory testing remains the domain of humans because it tests things that machines are less able to grasp.

The logic of the test pyramid remains valid. Much can be automated, and AI can take over entire test levels. In the end, however, tests should still be carried out exploratory and manually, supplemented by random checks of the AI results.

This is a common misunderstanding. In practice, exploratory testing is often used as a stopgap solution when you don’t want to test properly. In future, it will be about the opposite: consciously developing good exploratory testing and carrying it out in a structured, methodical way in order to find errors.

Perhaps this is actually an area in which we as humans will become even stronger. There are elements that the machine doesn’t test so well.

• Nicolas Nwabueze

How to prepare for this future today

Individual employees can adapt faster than entire companies, and that’s your opportunity. A company has to rebuild large processes. For you as an individual in the test area, switching to a new model is much easier.

The first lever is the skillset. The path from I-shape to T-shape to V-shape means learning new things, mastering new tools and delving deeper into related disciplines.

The second lever is attitude. A model without roles and hierarchy places high demands on self-organization and personal responsibility. This responsibility is no longer borne by one test manager alone, but by everyone in their own area. This does not suit everyone, and this is precisely where personal development pays off.

The third lever is dealing with AI. Many people were initially afraid of tools like ChatGPT. After a first real insight, this shifts noticeably and they dare to approach it. A practical introduction is where developers are reluctant to test, such as unit tests. Many are aware of the benefits, but very few actually use them. This is the next step.