Ekkono means *cognition*, and that’s what we add to the world of connected things (IoT). We make them smart through self-learning, anomaly detection, predictions of future faults, and by identifying what makes them work at an optimum and be more intuitive. The unique component, based on seven years of machine-learning research at the University of Borås, is a small embedded Advanced Analytics software engine that we license (OEM) to software and hardware developers. Ekkono’s tiny footprint enables it to run on the actual device, i.e. do Edge Analytics. This applies even to small micro controller platforms at par with RaspberryPi/Arduino, and smartphones (iPhone, Android). The result is personalized intelligence for ever individual unit, instant response, less data traffic, lower network load, and consistent functionality, even when losing connectivity to the Internet. This translates into money as you proactively avoid unscheduled maintenance (Predictive Maintenance), and by optimizing each asset’s performance. You upgrade your things from being connected to becoming smart. Ekkono’s solution increases data integrity by sending pre-processed instead of raw data, which is hard to interpret for a third party. It also makes your products more pleasant as they become more intuitive. By doing this on the actual device, you make better use of computing capacity; And by adding Ekkono’s crowdsourcing capabilities to cross-reference between multiple units, you make use of the intelligence of many.