Quality Assurance and Testing

In functional testing, the function of a product is checked to ensure that it does what is defined in the requirements and the corresponding acceptance criteria. Input is executed and processes are run through to see if the results correspond to the expected results.

The so-called end-to-end test is a technique to test whether processes within a digital product behave from beginning to end as expected from the viewpoint of the user. In this, several user stories are tested end-to-end with the goal of ascertaining system dependencies and ensuring data integrity between different systems. Interfaces, communication with other systems and databases are also included in the testing process.

An important component of the testing is the compatibility of an application or a website. The functionality and the design should work on different devices, browsers and with different operating systems and platform versions. Compatibility testing checks to see that the functionality behaves the same way or if deviations occur or if the function can’t be used at all. Since there are many different sizes and resolutions of screens, the design and layout should be checked to see how they behave on the mobile devices and browsers. The size of the display, the CPU and RAM also often varies depending on the device. Especially on older devices, errors can quickly crop up. Before development gets started, the target group should be clear so that the devices, operating systems, platform versions and browsers on which the application or website should run can be defined.

Would you like to know if your software runs perfectly even under increased use? Performance testing can be subdivided into two categories: load testing and stress testing. As the name suggests, in a load test we measure the performance of the system under a defined load (user requests). In doing so, a certain number of users (for example, 1000) is simulated accessing certain components of the software and executing actions at the same time. From this, we can analyse how applications and servers deal with an increased number of requests (for example at peak times) and if the system works perfectly under increased traffic loads. In contrast to the load test, a stress test tests the system with a maximal load to gain insight into potential optimisations of the system. The aim of a stress test is to stress the system with a very high number of users and to bring it slowly to a total breakdown. In doing so, we observe how the system reacts after the breakdown and if it can recover by itself. In addition, it is also possible to make a statement on the number of simultaneous interactions it takes to bring the system to a standstill.

With security tests, we check applications and systems for weaknesses and security gaps which could be exploited by hackers or unauthorised persons. We highlight potential threats and risks in order to avoid the loss of confidential information. Possible security tests are, for example, vulnerability scanning, penetration testing, risk assessment, security audits or ethical hacking.

We want to simplify the lives of our users. Usability signifies the user friendliness in the use of interaction systems. In usability tests, the effectiveness and efficiency as well as user satisfaction are measured and any confusing and obstructive issues analysed. “Real” users are invited for a usability test in order to review the products and systems for their usability. At an earlier stage of development, heuristic evaluations with usability experts are also suitable. Through usability testing, the vast majority of issues can be detected at an early stage and redesigned or made user friendly in a (more) cost-efficient way.

Automated testing is worthwhile for large and extensive projects in particular. Automated tests allow the expenses and efforts for regression tests and often repeated testing to be reduced. It is important to draw up an automation strategy regarding the software quality to be achieved before beginning with automation. It is often the case that the focus is too strongly centred on the GUI, so we generally recommend taking the automation levels of the test pyramid into consideration. Here, the majority of tests to be automated are at the lowest level (unit test) and second-lowest level (API and integration tests). Automation at the highest level (GUI level) should be kept to a minimum where possible, since the maintenance costs of the test scripts are extremely high and conducting the tests is very labor intensive and costly. In addition, the right framework/tool has to be used and the automation environment must be in place. The quality of a software can be increased dramatically using automation, but automations are very complex and are often underestimated. Don’t hesitate to ask any questions you might have. We’re happy to advise you.

Unit testing is a software testing method in which individual units of the source code (so-called "units") are tested to determine whether they are fully functional or not. Unit tests play a central role in smart contracts, since the source code cannot be changed afterwards.

The aim of in-the-wild testing is to use an application under real conditions. We test your application or website by putting ourselves in the position of the end user. Imagine, for example, you develop a sports app and the app is only tested in the office under laboratory conditions. We go out with your app and the accompanying sports watches and trackers into the natural, sporting environment, and we test out the features or put these through their paces in defined scenarios and test cases in order to scrutinise all functions under real-life conditions. The in-the-wild tests allows us to test the entire digital experience.