Many high-end celebrities and fashion models use personal shopping assistants or stylists to help find the best clothes for them.
For many women, however, that’s not really an option. We often rely on friends or family to give a thumbs up or thumbs down on potential clothing purchases, but that’s about the extent of it.
This may be about to change.
Lily is a virtual AI shopping assistant that learns about your style and how you feel in your clothes.
The app just won the 2017 SXSW Accelerator Pitch Event in the Social and Culture category. Here’s why it’s so much better than every other shopping app out there.
What Is Lily?
Put simply, Lily is a personalized mobile shopping app that helps shoppers discover clothes that suit them. For now, the app is only available on iOS (for free), but creators Purva Gupta and Sowmiya Chocka Narayanan promise an Android version is on the way.
Lily’s stated goal is to help women discover and buy clothes that make them look and feel their best. Created by an all-women team of six industry professionals with backgrounds ranging from Facebook to Macy’s, Lily has been shaped by the very audience it seeks to help.
“Lily is built by women, for empowering women to be the best version of themselves,” Gupta told me via email. Gupta says she and the rest of the Lily team have learned that there are 120 million U.S. women who are not happy with their appearance, and that women experience about 13 negative thoughts about their bodies every day. “As a team, we are on a mission to change these numbers and solve this problem in the world,” she says.
There are plenty of shopping apps on the market, but what makes Lily stand out is its technology. It doesn’t simply use your purchase history to give you bland or generic recommendations. Instead, it is designed to go much deeper, creating an experience that feels like the app is truly getting to know you.
But how does the app “know” what each woman wants in her clothing? That’s where some unique technology comes into play.
How Lily works
The Lily app is able to make suggestions and recommendations based on users’ perceptions and emotions about their own body.
It accomplishes this feat in a fun way. The app asks users a series of questions about body type and style preferences, almost like chatting or texting a friend. One example of a question Lily asks: “How would you describe your décolleté (that’s French for chest — I’m very worldly!)?”
On the surface it may feel like you’re chatting with some sort of automated quipster, but Lily is hard at work using your answers to learn more about what kinds of clothes make you feel your best. The app asks how you feel about your body — what parts you like to accentuate and which ones you’d prefer to minimize — and then uses a complex matching algorithm to make recommendations.
Lily’s creators call this algorithm the “Perception and Empathy engine,” the first such engine of its kind.
“I have personally spent more than 10,000 hours in the last three years asking women about what they feel when they are buying clothes online or in stores and why they buy the clothes they buy,” Gupta said when asked about the idea for the engine. “We quickly understood that most clothing recommendation engines are focusing on what users like and buy.”
Gupta says this information is obtained through “millions of tangible actions,” like browsing and past purchase history.
Through their thousands of conversations with women over recent years, the team at Lily learned that a true personalization engine needed to get beyond the what to the root of why customers buy the clothes they do.
Instead of focusing on tangible customer actions and rational behaviors, Gupta says Lily’s Perception and Empathy engine considers the intangible perceptions and irrational behaviors that drive consumers’ purchasing decisions.
“We also found in our research that women are buying more clothes every now and then to feel their best in the situation they are in or preparing for. It’s all about satisfying the feelings,” Gupta says.
The app connects the patterns of responses and generates suggestions that aim to please users’ emotional needs and style desires. It then gives you the option of buying clothing online, reserving items in physical stores, or taking the app along with you on a real-world shopping trip.
Lily has a number of well-known retailers on board already, including H&M, Express, Macy’s, Bloomingdale, Nordstrom, and Banana Republic. Gupta reports that one in three Lily users have already made purchases from their favorite brands through the Lily app.
Why Lily works
One of the reasons Lily is poised for success is because it’s a win-win proposition. Put another way, the app is valuable for its users as well as for the retailers that are on board.
For users, the app gives them unique suggestions and has a personal feel that can help them select clothes they’ll actually be happy with.
After all, the app “knows” millions of fashion rules/hacks that help it select clothing to flatter various parts of the body — tailored to your preferences and desires, of course. It also learns about you as you use it and prioritizes that information accordingly.
The app is also a win for retailers, for a couple of reasons.
First, it’s a good way to generate business. These days, retailers are looking to compete with online giants such as Amazon. In fact, Amazon sales account for over 60.5 percent of online sales growth.
Second, Lily attempts to provide users with very personal suggestions, which can lead to a good overall shopping experience. This helps paint retailers in a positive light, increasing the chance that Lily’s users will become return customers.
Why Lily won at SXSW
In short, Lily was honored at SXSW because it takes technology to the next level.
We know our electronic devices can “learn” our preferences, histories, and more over time, but very few applications feel this personal. What other app asks you which color you want to avoid more than kale fries?
On the business side, Lily already has a strong foundation of retailers on board. This helps in two ways. One, it gives the app instant credibility because it includes retailers that just about everybody has heard of. Also, it includes retailers that vary in price point and fashion sense, meaning there’s plenty of variety to go around.
And, assuming the rollout of Lily is as successful as it seems it will be, you can bet on even more of your favorite retailers jumping on board in the coming months.
And what does Gupta like most about Lily?
“My favorite thing about Lily is how she explains every item of clothing and how it flatters my body,” Gupta said. “We have built that feature ingesting more than 50 million data points, and I’m very excited about how this product feature has shaped up in the Lily experience.”
“I hear often from our users that they are getting spoilt by Lily as they now look for Lily’s personalized recommendation logic if they see any item of clothing elsewhere online or in-store.”
VentureBeatVentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact. Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:
- up-to-date information on the subjects of interest to you
- our newsletters
- gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
- networking features, and more