Most frequent testing mistakes your end users are going to suffer

Can you imagine buying a state-of-the-art car that, as soon as you touch it, the steering wheel breaks down in your hands? What would you think if, when you turn your TV on, it suddenly melts down? The answer is very simple: that it has not been properly tested, or, in other words, that the quality of the product has not been checked.

Applying this to how a TV app works, for example, you would be able to decide within seconds whether you like it or not. In a matter of minutes, you may have even  left a review (possibly negative) and who knows, in a matter of hours, you may have decided to unsubscribe from the service. This would all be caused by a wrong testing on that app.

Here you can find the most frequent mistakes (but not less important) that are usually made when you don’t have a good testing culture.

Not using the necessary resources:

In terms of effort: the validation of each project takes up approximately ⅓ of the total time of the process of developing a product/service. Therefore, it is essential to start validating the product from the design phase, since this will allow you to launch it earlier with greater security. In addition, the time available must be weighed against the number of people dedicated exclusively to a project, and the best possible way to meet these objectives must be defined, without delaying a process unnecessarily.

Regarding technologies: If it is the customer himself who provides the validation requirements, it is possible that some important detail was overlooked or not included in the documentation, and therefore, it would not be validated. For this reason, it is essential to have human teams specialized in the different testing models, with the specific tools for each type of test. On the other hand, one should not confuse the task of a developer with that of a QA, nor take for granted that the developer should be able to identify possible bugs since his environment is exclusively limited to code, not working in a real environment. A good QA should identify the source to which a problem may be pointing, and, with the information gathered, create synergy with the developer. In this sense, it is important to be able to do field trials, being this a much more adapted way of testing since it is done in analogous environments to those of the final user.

Not deepening enough in the diagnosis of a problem.

“Never leave to tomorrow what you can do today”. It is possible that the client does not have enough bandwidth to test certain functionalities, and that, even so, it launches a release where, inevitably, the first one to experience the failures will be the end user. Although it is true that it’s never possible to validate 100% of a project, for which you must prioritize a % of the most representative use cases of each functionality, thus ensuring its output to production. Depending on the time handled and the complexity of a possible unidentified bug, the QA must decide whether to do smoke testing or acceptance testing. 

Not making specific test per project and device.

Each project is a different, and has very specific requirements, by this we mean, above all, the backend. Each backend is different and each app launches the information requests in a different way, although the result of these orders is transparent to the end user. Not all TVs where you watch your favorite series receive the same information, each model or operating system can have many variations. Therefore, we must have a test case for each concept or device, although we can execute it as many times as necessary in the same, but not in other projects. That is why we say no to reusing test cases between different projects!

Not having a clear global methodology.

If the Agile methodology is here to stay, it is a sign that something is working well. With this, we do not mean that it is the only one, nor the most infallible, but we do consider it essential to have a clear and programmed working methodology. Currently, due to the supply and demand within the TV sector in relation to video on demand platforms, the evolution of the product is a constant process (or at least, it should be). Therefore, it is recommended to update the software on a recurring basis, with the aim of including new improvements or corrections with respect to the previous version or new functionalities that provide greater value to the product.

Not having both, automatic and manual testing:

You can’t put all your eggs in one basket. A QA is capable of performing many tests, but we must not forget that it’s a person, not a robot. Therefore, a very good practice to cover a wider range of validation, is to synergize the manual testing with the automatic testing, in stability validations, performance and user experience. That is to say, with types of tests that require many repetitions (minimum +100 iterations) or a long time of testing, the ideal is to automate. But the combination of both types of tests is crucial to ensure the quality of your service.

Consequences of the lack of testing in your service

Naturally, every action (or lack of it) has its consequences. We’ll tell you about them:

• Your budget will always end up with a few extra costs. You will include over costs by not detecting problems in time, which of course you will have to solve later, when you find a drop in sales, or dissatisfied users. 
• Your NPS will go down and you will lose the loyalty of your users or customers and, let us tell you that, when trust is lost, it is very difficult to regain it.
• You may receive a bad rating from the customer. It is easy to give visibility to this nowadays, since both the App Store and Google Play Store make it easy, like many other forums, to reflect a rating on an app in a matter of seconds.
• Churn and/or portability will be increased. Many customers will wave goodbye to you as they leave with your competitors.
• All this will result, in many cases, in a reduction (or no growth) of income for the organization. The balance between the investment made in product development, the lack of income and the over cost for the weak bet in testing, will undoubtedly be negative. 

Testing is an essential part of the development cycle of any TV software or hardware. That is why, if as a company, you make a bet on a product, you must bet everything on the winning horse and guarantee that the product meets the quality requirements, thus ensuring the satisfaction of the end user. Performing good testing is the key to success for any company that has a TV service.

What about you, do you want to give your customers the best quality and user experience? We have a team of highly qualified QA Engineers at your disposal that will provide you with the peace of mind you need and allow you to focus on what´s important.