When you’re looking for the best prices on flights and hotels for your next vacation, you might want to check out an app like Hopper. The company collects 750 billion prices each month, which now include flights and hotels, to analyze and predict prices so that you get a great deal on your trip.
Hopper’s machine learning algorithms can answer anything from “When should I book my flight?” to “Where should I go on my next trip?”
Mastering all of that personalized data-driven advice is Ella Alkalay Schreiber, Hopper’s Vice President of Data Science.
“What was once done by a human travel agent is now done through a machine that gets smarter each time an action is or is not taken,” Schreiber said.
To become the leading source on travel data and advice, Schreiber and her team have collected and analyzed trillions of price points. Hopper has distinguished itself in the travel industry with its unique data-driven conversational commerce, and the data science team is constantly working on building a richer and more dynamic dialogue with their customers. This conversation and user intent data is key to the company’s ability to implement algorithms and provide users personalized recommendations.
For example, earlier this year Hopper began testing a new recommendation algorithm which models meaningful insights into how much users are willing to pay extra over the lowest price for different flight qualitative variables like stops, cabin class, departure time, carrier etc. Every conversion strengthens the algorithm, thereby making future flight recommendations even more relevant for the specific customer, removing friction and empowering the customer consideration set.
Additionally, Schreiber and her team utilize the data to collaborate with the public relations team to serve as a trusted source for journalists when they’re working on travel stories. Her team also collaborates with airline and hotel partners to help them explore new strategies and opportunities based on Hopper’s unique set of demand and pricing data. In this way, Hopper’s data science team is delivering lasting impact for both business partners like airlines and hotels as well as travelers planning their next trip.
Schreiber first entered the field of data science in Israel, at Outbrain, the world's leading performance-driven discovery and native advertising platform. She was a data scientist in the recommendations group, where they served personalized content recommendations to over 500 million unique users. Her transition to Hopper from there was an organic one, as Outbrain’s recommender systems are similar to the algorithms Hopper serves. At any given moment, there are thousands of potential recommendations Hopper could be offering a single user. The key to success is reaching the right user with perfect flight recommendation at the right time.
“As a data scientist, it’s important for me to work in companies where the value proposition is the data,” Schreiber said. “I wanted a company that didn’t compromise, and strives to be the best in that field.”
When she began at Hopper as a data scientist in 2016, the team consisted of three people. Now, Schreiber manages a team of 20 within the 300-person company.
Schreiber is committed to ensuring that her team retains its communication, processes, and impact as it and Hopper continue to grow. They’re currently working on implementing new machine learning frameworks to help support both the flights and hotels side of the business. With these advancements, Hopper can extend and improve its conversation in the company’s signature user-centric and data-driven style.
“Hopper is always evolving,” Schreiber said. “The opportunities and challenges are huge, and the more we grow, understand the industry, and collaborate with users and airlines, the more interesting our environment becomes.”
Quick Q(uestions) and A(dvice)
What do you enjoy doing in your free time?
I love going diving -- that’s usually my preferred type of vacation. I scuba dive all over the world. I studied 10 years ago on an island in Honduras, and since then I’ve spent almost every vacation in a diving site.
How do you manage stress?
I focus on causality and action items. I think understanding the root cause of the stress and talking to my team helps me deal with it. Taking time off is good, too! It’s important to have an escape.
How many cups of coffee do you have in a day?
Too many! I think four cups a day now. We have a really good coffee machine here.
What’s one of your favorite places in the Boston area?
I love the Esplanade. There’s a good beer garden along there to have a drink, watch the water, and relax. It’s also dog-friendly, which is perfect.
What do you consider one of your greatest accomplishments?
Establishing this team at Hopper and building it from a small startup to where it is now has been my proudest so far. Finding amazing people in this space has taught me so much. We’re looking for people who are inherently curious and are exceptional problem solvers. Everyone on our team is very independent, so we’re looking for people who are self-sufficient and ready to take on and own challenges. We spend a lot of time collaborating with other teams. In addition to having the right technical skills, members of the data science team must have strong product intuition, business judgment, and the ability to articulate their findings to non-technical people.
How does where you are now compare to where you saw yourself 10 years ago?
I didn’t know I’d move to Boston and lead a team like this! Every day at Hopper is a fascinating day -- there are so many ideas, projects, challenges, and new hires. Every day I feel fortunate that this is my job.
What’s your advice for recent college graduates?
Make sure you work in your passion. Work in something that you think about all the time. When I’m hiring, I’m always looking for people that have passion, curiosity, problem-solving skills, and good communication. If you have all that, you’re unstoppable.
