Or listen in your favorite podcast app
The old saying, “look good, feel good,” fits Stitch Fix perfectly. The direct-to-consumer, online personal styling service has boomed due to its ability to not only match consumers with trendy and comfortable clothes, but to make it a personalized experience for each buyer.
“At the end of the day, we want to be rooted in personalization. First and foremost, it’s important for us that [consumers] not only come to Stitch Fix to complete a transaction, but that you really think about us as your partner style.”
From curating stylish experiences, to matching each individual with their own personal stylist, Stitch Fix and taken eCommerce and the subscription model to the next level. On this episode of IT Visionaries, Tatsiana Maskalevich, the Director of Data Science at Stitch Fix, takes listeners into the algorithms that help curate those experiences and the science behind it all.
- Know Your Data: With so many avenues to collect data on the internet, the collection process is not the difficult part. The hardest part is how you turn your data into actionable insights that can benefit your customers’ experience. In order to manage your data appropriately, you must be constantly creating feedback loops so you can best understand what is working and what is not working.
- It’s a Preference Thing: The customer experience is the number one pillar for any successful company. When you are creating curated experiences for your customers, you need to constantly be listening to what works for them and what does not work. Create a great experience by constantly giving your customers reasons to engage with your platform through quizzes and games or other interactive features.
- Rooted in Data: One of the keys to building a successful data-centric platform is through the individuals constructing it. The algorithm you construct has to accomplish two things: It needs to continuously be learning about the person inputting that data, and it needs to continuously be relaying information back to the source. Focus on the foundation and infrastructure of your algorithm first, then experience second.
For a more in-depth look at this episode, check out the article below.
The old saying, “look good, feel good,” fits Stitch Fix perfectly. The direct-to-consumer, online personal styling service has boomed due to its ability to match consumers with trendy and comfortable clothes, but to make it a personalized experience for each buyer.
“At the end of the day, we want to be rooted in personalization. First and foremost, it’s important for us that [consumers] not only come to stitch Fix to complete a transaction, but that you really think about us as your partner style.”
From curating stylish experiences to matching each individual with their own personal stylist, Stitch Fix and taken eCommerce and the subscription model to the next level. In this episode of IT Visionaries, Tatsiana Maskalevich, the Director of Data Science at Stitch Fix, dives into the algorithms that help curate those experiences and the science behind it all.
Launched in 2011, Stitch Fix provides consumers with the ability to find their perfect style and fit, all within the comfort of their own homes. The process is relatively simple: potential clients are asked to fill out a list of questionnaires centered around which brands they often wear. From there the platform helps narrow down the size and fit, and the company then matches each potential customer with their own stylist who then creates a curated list of packages. It’s a process that sounds simple, but with thousands of brands with untold amounts of products, sizes, and styles in its inventory, Stitch Fix relies on data more than anything else.
“Data is the key to a lot of things at Stitch Fix,” Maskalevich said. “We’re quite lucky. Not only [do] we have a lot of data, we do have a very deep knowledge of our clients because when we set on a journey to help you feel and look your best, we also created a lot of feedback loops.”
Those feedback loops are integral to the success of Stitch Fix. Especially during the early stages of when a customer signs up, through the selection of products, the Stitch Fix system is constantly learning and understanding what products best fit a customers’ needs.
“You can imagine that we have this perfect feedback loop that gives and [helps understand] how each garment is constructed, but we also know exactly what [a customers] preferences are,” she said. “We get the connection, the data point that allows us to really make an inference about how each piece of clothing is going to work for each individual consumer.”
So how does Stitch Fix match you with clothes that best work for you, but also pair you with a stylist that fits your needs? It’s a process that begins from the moment a customer enters the funnel to the end of their journey.
“[We] construct something we call a lead and style map,” Maskalevich said. “We can also place our stylist on the same map because just as consumers reveal their preferences by ratings, our stylists rebuild their style, based on what they like to send to clients. While they’re serving a wide variety of clients, each of us have our own favorites and our own preferences when it comes to ascetic. So as you could imagine, we can place clients’ preferences in that multidimensional space. And then we can put our stylist preferences in the same space and it’s a matter of mathematical function.”
That function is part of the engine that fuels Stitch Fix’s data model and, according to Maskalevich, Stitch Fix collects billions of data points across their platform. From its Style Shuffle game, which helps the system learn what looks consumers enjoy through their engagement, to understanding which articles of clothing often get returned, it’s a process that can be overwhelming, but the key is understanding how to best utilize the billions of data points they are afforded.
“When we think about the data science of Stitch Fix, it comes in multiple layers,” Maskalevich said. “We want to have our scientists focus on science. So building those algorithms that derive the insights from the data, and not just in the form of communication, but really in the form of the matching and in the form of systems that we put in front of stylists and clients, both from the experience point of view, but also the recommendations themselves. The secondary piece of that, that as a team, we really focus on really laying a good foundation for our computer infrastructure and data management infrastructure.”
“Choice anxiety is a real thing,” Maskalevich said. “We have to make so many choices and so many decisions every hour. We have thousands of items in our inventory, but we decided to only show you things that we think are the most relevant for you. So in that sense, you can come back and look at it as we refresh them. But we are very much a curated experience even when we’re serving a lot of inventory.”
That curated experience for the team at Stitch Fix is important to the company’s success. As more-and-more individuals embark on their own style journey, the company wants to make sure that they are viewed as a viable option to brick-and-mortar stores, but as a partner in their customers’ journey.
“At the end of the day, we want to be rooted in personalization,” Maskalevich said. “First and foremost, it’s important for us, [consumers] not only that you come to Stitch Fix to complete a transaction, but that you really think about us as your partner style. There are different factors we want. We want to see that you have a connection with your stylist. We want you to see that you have the connection and you really truly love what you get. So there’s multiple variables that go into [the success of the platform]. I wouldn’t say that when we look at evaluating the impact of the algorithm, we just look at the transaction and we will look at this as the whole picture.”
To hear more about how StitchFix is creating a curated experience through the power of data, check out the full episode of IT Visionaries.
To hear the entire discussion, tune into IT Visionaries here.