The problem
In many organizations, there is a strong and familiar belief: enterprise-grade testing requires enterprise-grade tools. As a result, high licensing costs are often accepted as a given, rather than questioned.
This mindset creates dependency on external platforms and limits flexibility in how testing processes are designed. It also discourages exploration of alternatives. Our experiment challenged this assumption by taking a different approach, using small, focused proof of concepts to validate whether replacing such tools is not only possible, but practical.
The Research Question
Our AI experiment focused on two closely related questions:
Can Lovable be used to build an AI-driven proof of concept that replaces core Testmo functionalities?
How effectively can AI support test management while reducing licensing costs?
The experiment
To explore this, we used Lovable (an AI tool that generates code based on prompts) to rapidly build a first version of a custom testing solution. Within a very short timeframe, we created a working proof of concept that replicated essential Testmo capabilities.
At the same time, Infodation had already developed its own software to collect test results from various testing platforms. By combining these efforts, we were able to replace the existing third-party solution with our own internally built tool.
This not only reduced monthly licensing costs, but also improved data control and security by keeping all information within our own environment.
Results
One insight stood out immediately: Speed and cost can coexist.
Using Lovable, we were able to rapidly build a functional POC that covered essential Testmo capabilities, at a fraction of the cost. What traditionally requires heavy tooling and long setup times could now be prototyped quickly, without large upfront investments.
What This Means for Our Work
These findings suggest a meaningful shift in how we think about testing tools. We can reduce our dependency on expensive third-party platforms and regain control over our test processes. AI-based solutions offer flexibility: tools can be shaped around our workflows, rather than forcing teams to adapt to rigid systems.
Perhaps most importantly, this experiment shows that when AI is applied thoughtfully, lower cost does not automatically mean lower quality.
The implications extend beyond test management. Internal, AI-powered tools can be developed to match our exact QA needs. Lovable-based solutions can be scaled to other internal processes. And the savings generated can be reinvested into quality improvements, innovation, and the things that truly add value.
Breaking the Testing Status Quo
This experiment shows that test management does not have to be tied to expensive, one-size-fits-all tools. By combining AI with smart design, we demonstrated that core testing functionality can be rebuilt faster, at lower cost, and with greater flexibility, without sacrificing quality. Small experiments like this help us challenge assumptions, regain control over our tooling, and make more intentional choices about where we invest next.
