Lewis Prescott: “We Throw Testing Parties and Monitor Edge Case Users”

How QA in Healthcare Can Make a Difference, Keeping a Human Face in the AI-Obsessed World

Lewis Prescott is a QA Lead at Cera Care, a technology-enabled home care provider. As one of Europe’s fastest-growing companies, Cera Care redefines healthcare by moving more and more services out of hospitals and into people’s own homes.

We’ve talked about the challenges of QA for apps used by non-tech-savvy people, the skills and qualities efficient QA specialists should have in this rise-of-the-AI era, and the benefits of “testing parties.” Parties, yes, you’ve read it right. 

Please, Tell Us More About Your Company and Its Impact on the Industry

The home care industry has been neglected for a long time, and as a result, it lags even further behind other sectors in terms of technology adoption. Our company’s mission is to transform the industry by automating and digitizing it. Many non-technical home care professionals are caught between the practical and administrative aspects of their job, and we see it as our responsibility to empower them with technology solutions enabling them to deliver their service more efficiently.

One major challenge we’ve encountered is the high level of paper-based documentation that accompanies home healthcare in the UK. When attempting to digitize the process, we want to ensure that caregivers spend less time on administrative tasks.

What Are the Biggest Challenges for You Right Now?

Our users are often non-tech-savvy, so if the app crashes, there’s not much they can do about it. They may not even feel comfortable restarting the app, which means the better the user experience, the fewer “this app doesn’t work” requests managers receive. 

As a startup, we want to release fast and add new features to stand out in the market. However, as a company that promotes an app for non-technical users, we strive to guarantee a glitch-free user experience. This creates an environment that requires a united effort from our team to put in the work for each function and release.

That is why QA plays such an important role, and that is why I coach our teams to work closely with QA engineers — if we want to have it all, we have to put a united effort into every feature and every release to create an excellent user experience. 

How Do You Address QA Challenges?

Our QA engineers work closely with developers to think about what we need in terms of automation, user acceptance levels, and other scenarios that help us have confidence in test coverage.

We use location monitoring through Firebase to drill down into issues, such as those related to poor network areas, which can result in issues being wrongly attributed to the app. It’s important to have an understanding of interactions with data to address the truly important issues. 

We also pay attention to real-life scenarios and think about how non-tech-savvy people would interact with our app. 

How Is Your QA Process Organized? Are There Any Tricks You Can Share?

Regarding our QA process, we have a unique approach that we call a “testing party” (and everyone is welcome!), introduced by a colleague of mine. It is similar to Mob Testing, a group approach to testing where different roles get involved and offer their different perspectives. We do it at the end of each sprint, where everyone involved in the project, including developers, product owners, and anyone else interested, participates in testing. 

Our goal is to get everyone on board with quality, so we emphasize the importance of testing for everyone on the team. By involving different viewpoints, we can ensure that there is less to no bias in the testing process. The result is a more comprehensive testing process that leads to a better final product.

So Despite All the “Shift Left” Mania, You Test Closer to the End?

Testing can be done at any point, but it’s important to find a balance between testing early (“shift left” approach) and testing thoroughly. Testing earlier can help catch issues before they become bigger problems, but testing closer to the end can provide a more comprehensive view of the product from the user’s perspective.

Ultimately, it’s important to focus on getting everyone on board with the QA process and making sure that quality is a shared responsibility. With a team effort, you can continuously improve the QA process and ensure that your product is meeting the needs of your users.

And How Do You Balance Speed and Thorough Testing?

It’s important to strike a balance between speed and thorough testing, and the approach may differ based on the specific feature or functionality being developed. To achieve this balance, I consider the risk and impact of the problem, the frequency of occurrence, and whether the feature needs to be delivered immediately or can wait.

In addition, we refer back to live data that we monitor and ask ourselves which feature will have the most significant impact. Once we fix a problem, we determine what the next steps are and prioritize testing journeys multiple times to ensure we catch any issues that may arise.

What About Automation? Can It Be a Silver Bullet for Speed/Quality Balance in QA?

No, not necessarily. At my company, we put a lot of emphasis on developers writing automation from the start, so we don’t have long-running end-to-end tests that need constant maintenance. It’s still automation, but it’s not from the user’s perspective. The emphasis is on coaching the testers to think about the user and what’s important to test. It’s more about collaborating with developers to cover important scenarios. This approach helps us reduce the manual effort during regression testing and increase confidence in our releases.

The human touch is crucial because the user experience is abstract, and metrics don’t capture everything. We care about feedback from our users and the complex negative cases that automation may miss. The focus should be on what is risky and important to test rather than testing everything.

Will AI Take Over QA? Should We “Brace for Impact,” So To Speak?

There is still a need for human decision-making and prioritization, especially in healthcare, with all the regulations and data that need to be verified by doctors and nurses. While AI can help with the speed and integration of data sets, it is not a major concern for job displacement in the testing space.

AI is definitely helping with the speed we can integrate data sets and reach the point at which we include a human. As testers, we have a creative element that involves understanding the user experience and thinking about negative cases that automation may not cover. Metrics and automation are important, but they’re not everything. 

QA can use the tools available (with or without AI) to facilitate their testing. The overall focus should be on the user, their experience, and how your product matches their needs and expectations, and how it can exceed those. A robot cannot do this completely, as it may lack context even with a large amount of data fed to it. 

Good To Know, Humans Are Still in the Game. Which Skills Do You Look For in QA Specialists When Hiring?

When looking for a QA engineer, I want them to be problem solvers above all. It is not about knowing the process. It is about being proactive toward the solution. They need to have a mindset of “I don’t know the answer, but I will find it.” 

Curiosity and a user-oriented approach would be the second. A specialist should be constantly contemplating: Which questions do I need to ask? Who do I need to get involved in this? What do I need to learn in order to be able to solve my own problem? They should be self-starters who take the initiative to learn new skills and technologies and who always strive to improve their own performance.

Last but not least, I look for people who are great team players. We don’t bet on lone-wolf experts, and we don’t look for superheroes. Our “mob testing” concept is the essence of it. 

Does QA Outsourcing Fit in This Concept? Do You Have Such Experience?

Outsourcing QA can bring both benefits and challenges. In a previous project, we worked with an agency that had testers in South America and Europe, which allowed us to cover almost all time zones. This was a big time saver, as developers could hand over work to testers and then go to sleep, and work would continue overnight. We also found it valuable to have local access to testers who could help us test edge cases and latency in multinational, multi-language applications.

Another benefit we found was being able to tap into the external talent resources of outsourcing companies to fill our own experience gaps. For example, when we didn’t have mobile testing specialists on our team, we were able to leverage the expertise of an outsourcing partner.

Challenges we faced naturally came from communicating with ESL specialists. Also, simple things like sharing calendars and factoring in time differences into organizing team meetings were not always easy. However, we were determined to make sure everyone feels a part of the team.

All said I want to emphasize the importance of a balanced approach to testing, a necessity to always consider the problem’s risk, impact, and frequency of occurrence. Cold-headed prioritization and team-based testing give you more than obsessing over automation. Despite the AI-related frenzy, you should never underestimate the human touch in QA, as user experience is abstract, and metrics don’t capture everything.