Dogfooding is a slang term in the world of IT. It is a favored practice of companies to use their products and services.
It allows testing the finished product that, according to the QA team, meets the client's requirements in a real environment.
The history of this term is most often associated with Microsoft, which encouraged employees to use their software, and Amazon, where they had no alternative (well, if they wanted to continue to work there).
Why are closed Alpha and Beta testing organized for many games?
It is impossible to predict and test all possible cases because each person has a unique user experience that is difficult to replicate in a "laboratory" environment.
Therefore, after a professional QA completes the testing, the companies often give their employees their product for a test drive. After all, nothing motivates us to improve and optimize than the true experience. How often do you see an app on the AppStore that got a bunch of similar reviews on some trivial flaw, after which the users promise to put 5 stars?
Here's a real-life story that will help you understand why it can be so important.
It was decided at the office to reorganize the purchase of delicacies for coffee breaks. We did a poll: everyone does not care what to drink, and they are fine with how things are at the moment. We ordered several types of tea and coffee at our discretion. After that, within a month, we learned that one coffee brand was too strong, and the previous one had a strange taste, and in the end, everyone liked decaf (only they did not know that it was decaf). The same goes for teas: "any is okay" but in fact, groups of connoisseurs of earl gray and Ruby Girls who adore flavored black tea with added dried fruits were exposed. As for the herbal tea, it was tried once and left alone and forgotten in the far corner of the locker for half a year.
"Overall, everything is okay" is a highly insidious and erroneous generalization. Personal experience allows one to evaluate a specific thing or service subjectively, look at the little things from a different angle, and achieve great results before the actual production.
Of course, there are a few "buts" here:
- Everyone on the team should take this seriously.
- and should keep in mind that the sample may not be representative because not all are potential users of specialized services.
Even so, this is a very good start which allows the team to understand the end-user better, gaining their own experience of using the service they worked on.