The most valuable would be their Live Stream analysis, where I can see the live analysis of all the executions on a single device or multiple devices as well as track them. The live analysis and reporting would be the single most valuable feature. We leverage Perfecto’s reporting and analytics a lot. From the CI Dashboard, it is mainly the status, which is the past, failure count, and time consumption, e.g., how much time did an average test or script take? Along with that, it provides the historical view compared to the previous result, e.g., am I a pass or fail? Also, the stack trace is very important. Whenever a pass occurs, we don't look beyond that. However, whenever a failure occurs, the stack trace information that it gives us is pretty critical for us when figuring out where failures lie. It gives a summary for the pass/fail count, total test count, the historical view, time consumption for each test as well as the total tests, and the stack rate of the failure. Perfecto's analytics are very important since we use them on a daily basis. We run our executions daily. After every execution, we pull information from the Perfecto reporting system and share that with our stakeholders. Having this information accurately reported is pretty important for us, so everybody is aware of the current status of the product. That way, we can evaluate the health of the product or environment against that which has been executed. Therefore, it helps make those real-time decisions and highlights the impact to the business. I found Perfecto to be pretty easy to use while executing against cross-platforms. The main reason is because the same script or test automation where we execute on multiple platforms has minimal changes that I need to do. Also, it is easy for me to set up an execution on one platform, then on another platform, either in parallel or one after the other. Parallel opportunities save me time. Once the execution has been completed across these different configurations, I can always check and compare, e.g., what are the differences and consistencies? We utilize Perfecto’s cloud-based lab to test across devices, browsers, and OSs. I use it occasionally for manual testing. Though, there are other team members who use it more frequently than I do. I use it mainly for executing my automated tests. We have the Perfecto lab, cloud devices, and machines. I can program my test to execute against any of those devices, which gives me more confidence in my product. I can compare and see how my product or application functionally behaves across these different devices and from a UI point of view, which helps me a lot. The device lab is extremely important to our testing operations. We rely on having multiple devices up and running all the time. Whenever we kick off an execution, there are multiple reasons why executions may get triggered: * CodeCommit * A scheduled job. * Might be on-demand by any stakeholder. We need the lab to be available, as we need devices up and running for executions to take place. Also, the devices help since they allow us to have parallel execution, and not just wait for a sequential device to become free and available. Therefore, volume is definitely key. It also gives us an opportunity to compare execution across platforms in that space. It is extremely important to you that the lab provides same-day access to new devices since we analyze that data every single day after execution. Perfecto provides their own framework called Quantum Framework. That is one option. The other option is, if I want to have my own framework, I can have a Java-based Maven project, take a Selenium library, AppiumLibrary, and REST Assured library, and utilize the open-source framework. It is easy for us to connect to Perfecto, no matter what framework we use, as long as it has these core libraries in it. I can design and structure it any way that I want. The execution will happen in Perfecto no matter what since they have support for these tools or libraries. It is pretty neat that way. We are not dependent on using just one particular framework to use Perfecto. While there are still some framework limitations, there is the opportunity to use multiple, different open-source frameworks, then pass the execution to Perfecto. We can use most frameworks, then design and craft it any way that we want, then just pass the execution to Perfecto.
