1. As a platform, we can take advantage of the network effect. The data we gather from all the users using it, plus the interaction they do when we classify certain patterns can be used to improve our machine learning algorithm. Thus, you as an IoT manufacturer can take advantage of a trained algorithm instead of you developing one from scratch.
2. This network effect brings in another discussion which is privacy and security, we are very aware about this. The data is never shared with third parties and it's only used for training/improving our algorithms, we protect user data, because we know this is a big concern that will surge with IoT in the future. We are planning to implement end-to-end encryption and open-source the core code at some point so you can deploy your own local applications.