A successful product is a result of a huge number of trials, errors and a bit of luck. However, if you don’t test ideas and new product features, you don’t know if your decision was mistaken or not and you miss a chance to get a valuable lesson showing you the right direction.
While we are exploring other products and their features for our competitive analysis, these companies are most likely running experiments
If you work in a product design or UX field, you are probably familiar with competitive analysis. Often while creating a product we look at the most successful products in a niche as references. The thing is, while we are exploring other products and their features, these companies are most likely running experiments. There is a big chance that we explore experimental features, and blindly implementing something similar to our product is a huge risk. Experiment driven is not a process, but rather is a company’s culture.
In this short article, we will review different product stages based on the annual number of experiments.
You are moving slowly to your product Olympus. The less you experiment, the less you know your users. That also means you often make wrong product decisions that teach you nothing and wasting your time. This is the reason why you grow so slowly.
You start realizing what your audience is sensitive to, and what your users not reacting to. You know how to validate your metrics through experiments, but you are still slow and doing a lot at the level of experiment representation. However, you are definitely on the right track.
You definitely know how to get in with your users, but, it doesn’t stop you, and you do everything to experiment more and more. After all, experimentation is one of the reasons why your product and business are growing.
This is a quite relevant classification for modern tech companies. Though, it is important to understand that high qualified team and data-oriented product management are crucial ingredients of a successful experiment-driven culture. Also, solid data architecture, which allows running a large number of experiments and an automated reporting system are very important too, but this is a topic for another article.