Blog

October 9, 2018

Atolla: Skincare Powered by Machine Learning

Atolla is a company that delivers customized skincare products to its customers using a machine learning-driven platform. The company was founded by Meghan Maupin and Sid Salvi late last year, and the company was recently a part of MIT’s delta v summer accelerator.

I had the chance to speak to Maupin (also CEO) about how such an interesting company gets built, as well as the story behind its founding (and where it’s going next).


Starting things off, I'd love to hear the story of how Atolla was founded.

During grad school, I got sick and my doctor recommended I keep a food diary. In the process, I noticed my skin started getting itchy rashes when I ate certain foods. But I couldn’t be sure. My food diary turned into an obsessive skin diary where I wrote down everything I ate, how much I slept, what skincare products I used, what the weather was, and every detail you could imagine. I realized, wait, I’m at MIT, the best school in the world to solve this problem using data analytics and machine learning. And when I discovered how many other people had the same problem of not being able to figure out what’s happening with their skin (at least 2 million), I set out to find a technical co-founder. I met Sid at an MIT entrepreneurship event. He has eczema, but his background is analytics, so it was the perfect pairing of professional experience and solving a personal problem.

Your company is super interesting to me because it combines a machine learning software platform with actual physical skincare products. What was the process of building a company like that, and how does it work on your end?

My background is in experience design, and Atolla first started through an extensive user research project. We wanted to understand why people couldn’t solve their skin problems and buy the right products. The root cause of not being able to find skincare products that work is that people don’t understand their skin. There are so many factors that impact skin, our largest organ, and skin is constantly changing. So we designed a way to help people figure out what’s causing their skin issues by identifying the factors that impact their skin and measure them over time.

To further the skin science and classification, we brought Dr. Ranella Hirsch, an amazing dermatologist with extensive experience in product development, onto the team. We filed a provisional patent on the process and algorithm of measuring the factors that impact someone’s skin and matching people, based on their skin data, to the most effective personalized ingredient combination for them. We actually measure how well the formulation works and use the feedback to improve the model.

Atolla today delivers adaptive skincare by predicting what works for your skin using machine learning. We’re building the world’s first model to predict skin outcomes based on someone’s skin attributes and the products they use. We empower the customer by both identifying the cause of their skin issues and offering them a personalized solution. You can use our skin health tracking platform without using our products, but we get better insights if you do. Think of it as a personal skin experiment; the more we can measure, the more we can learn.

Meghan Maupin Atolla
Atolla Co-Founder and CEO Meghan Maupin.

Walk me through the consumer experience of using Atolla. If I wanted to improve my skin through your platform, what would that be like on my end?

After a successful pilot of the in-person Atolla Skin Lab, we’re excited to say we are going from in-store into in-home. The process is pretty easy and takes about 10 minutes. First, you sign up to get the Atolla Skincare system on our website (you can pre-order today). Then, you get an initial assessment kit in the mail with tools to measure your physical skin attributes like oil, moisture, and pH. You don’t have to send anything back to us! Everything is through the digital app. You use the paper-based strips to swipe your skin, it changes a color based on your measurements, and you take a picture of it in the app. Along with answering a couple of questions and taking a selfie, we’re able to “sequence” your skin- measure it across multiple attributes and compare to other people with skin similar to yours.

Following the initial diagnostic, you get a personalized serum and tools monthly to see how your skin is changing. We measure what factors (environment, diet, lifestyle, ingredients, etc) have the biggest impact on your skin health and determine what the most effective formulation is going to be, no matter the situation. On the back-end, it’s pretty complicated (which is why we are using machine learning), but for the customer, it’s simple and only requires 10 minutes a month.

How has the company evolved since its founding?

The more people on the platform, the more we learn. At the beginning of Atolla, we were definitely gathering a lot of data and asking a lot of questions. We’ve learned over time what the right data is to collect in order to get the answers we want, so we require less time and effort from the customer. We’ve also been doing this for about a year now, so we’re able to see some really cool trends like seasonal changes and geographic differences in people’s skin!

Overall, the process has gotten a lot more focused, driven by our mission to empower people to take ownership of their skin. We changed several things about our business model after having the Skin Lab open this summer, such as transitioning from a seasonal subscription to a monthly subscription, as well as our product offering. In the Skin Lab, we mixed custom facial oils because it’s a versatile product that’s easy to formulate on site. The Atolla Skincare System is launching with a serum. We took the knowledge we collected about people’s skin, ingredient efficacy, and preferences to develop the serums, and chose that product because it will create a transformational change.

We’re constantly evolving because that’s what we set out to do: use the data we collect to adapt in order to identify patterns and solve skin issues. So I’m excited to see what we continue to learn over this next year!

What do you notice the biggest differences and similarities are between the Boston and NYC tech scenes?

We’ve been fortunate enough to be part of both Boston and NYC tech scenes through our time at MIT and now in New York full time. It was great for us to incubate our technology while at MIT, with the technical mentorship of some of the best engineers in the world. This summer in MIT’s delta v accelerator, we really bridged the gap and started testing in a more real-world scenario. Atolla is in a new “beauty tech” space; our model is more like 23andMe or Fitbit than it is traditional beauty. So the great thing about New York is access to mentors and other entrepreneurs who have built strong DTC brands, especially around tracking health outcomes. Boston isn’t known for having a lot of consumer brands.

The focus of the two cities is definitely different; Boston is more heads down, more lab-work and New York is definitely a get-out-and-test mentality. We’ve found a great home working at New Lab in the Brooklyn Navy Yard with a lot of other companies working in advanced technologies- robotics, AI, ML, manufacturing, life sciences. It’s been great to be part of this community, as we are all building the future together!

What comes next?

We start shipping the Atolla Skincare System in Q1 of next year. After that, we’re planning on launching our next product and building out our insights platform. And we’re hiring!


Alexander Culafi is a Staff Writer at VentureFizz. He also edits and produces The VentureFizz Podcast. Follow him on Twitter: @culafia.