Smart and connected shavers for Philips
Recognizing the right accessory for the best shaving advice
Smart consumer products require reliable algorithms. For a new generation of connected shavers, Philips developed an application that automatically recognizes which accessory is being used and immediately provides the user with personalized shaving advice.
Together with CQM, a robust algorithm was developed that reliably detects accessories. This ensures the user always receives advice that matches the shaving head, trimmer, or cleaning brush being used.
Smarter
personal shaving advice
Reliable
accessory recognition
Less
risk of incorrect advice
From user needs to smart functionality
To find out everything about this project, we spoke to Erik Paauw, project leader for pre-development processes at Philips.
"Our category - Grooming & Beauty - develops consumer products. It is important to always start with the starting point: the customer's needs. In order to then be able to meet them as best as possible. In this specific case, we have spent over a year developing a robust algorithm for shavers. The men in our target group are always looking for a way to shave as close as possible (smooth, dry and wet), with as little skin irritation as possible. To meet this need, we have developed a new, intelligent shaver with a smart app that helps them by providing the right feedback.”
Reliable detection with algorithms
“First of all, the shaver recognizes which accessory (shaving head, brush, or beard styler) is being used. Vibrations and current are measured inside the device (the handle), and an algorithm determines which accessory the man in question is using. Based on this, the user is coached toward the best result through feedback via the handle and the app! For example, it provides advice regarding the correct shaving technique. About the pressure applied, whether it should be higher or lower, which shaving stroke is best, and so on. Incorrect detection therefore leads to incorrect advice! And thus to a less smooth shave, more skin irritation, or an imperfectly trimmed beard. We absolutely do not want that, and CQM has helped us achieve error-free accessory detection. In short: by recognizing the accessory – unit recognition – and providing the right advice, we can now offer a personalized shaver. If desired, we coach the user step by step through their shaving process.”
Developed for long-term use
"We have also taken into account all kinds of developments that may occur over time while using the device. Consider, for example: what effect does wear and tear of the device or the life of the battery have on this coaching? The application must of course provide specific and appropriate coaching not only now but also in the future. To avoid any surprises, CQM has included some risk assessment so that the algorithm works as accurately as possible and is future-proof."
Robust testing and validation
I really enjoy working on these kinds of pre-development projects. It starts with mapping out customer needs, conducting research, understanding the struggles they face, and truly getting to know the customer. These are long-term projects, but the results are often stunning. Statistics are very important in such a preliminary phase. You need to know what effect something is going to have, and the tests you conduct must ultimately apply to a larger target audience. Before a product is launched, everything has to be right down to the deepest detail. That means a lot of testing, removing all the bugs, and gathering a lot of knowledge through trial and error. It is good to have an external party assisting us throughout that entire testing period.
From algorithm to product
“After thousands of measurements and multiple verifications, we were ready for the launch halfway through this year! And it was a success. CQM helped us detect very accurately which accessory is on the shaver, and then translate this into ‘okay, I have a brush, but what is the best way to use it now?’. The goal is for every user to receive the right advice. Launching a smart product means continuous development. Together with CQM, we have made a robust start and we will constantly improve to offer the best to our consumers time and again. The other department within Philips, which CQM also supported, will help us enormously with this: analyzing Reviews & Ratings to understand the customer even better and improve our products.”
Want to develop smart products with reliable algorithms too?
CQM helps organizations with algorithms, statistical validation, and data science for innovative products. Contact Emy and discover what data science can mean for your product development.