KDnuggets Home » News » 2014 » Dec » Opinions, Interviews, Reports » Interview: Igor Elbert, Gilt on Boosting Sales through Analytics-curated Shopping ( 16:n16 )

Interview: Igor Elbert, Gilt on Boosting Sales through Analytics-curated Shopping


We discuss Analytics at Gilt, unique Analytics challenges of a flash sales portal, consumer behavior across channels, interesting insights, advice and more.



Igor ElbertIgor Elbert has been dealing with big data for over 20 years. From calculating financial risk for Salomon Brothers to tracking movements of millions of items across the supply chain for major brands, Mr. Elbert pushed innovative data analysis to new frontiers.

As VP of Quantitative Analytics for Barnes & Noble Mr. Elbert used a plethora of data to offer his customers a unique in-store and digital experience.

Having joined Gilt.com as Principal Data Scientist, Mr. Elbert is supporting Gilt’s mission to create the most exciting, curated shopping experience that helps company’s customers find and express their style.

Here is my interview with him:

Anmol Rajpurohit: Q1. What does Gilt do? What role does Analytics play in it?

Igor Elbert: Gilt_LogoGilt is a membership-based online retailer that provides instant insider access to top designer labels. Our members find something new every day for women, men, kids and home as well as exclusive local services and experiences. Analytics plays a key role in understanding customer preferences and behavior. Analytics also helps with merchandizing, pricing, measuring and optimizing sales performance, etc.

AR: Q2.  How is Gilt using Big Data to deliver personalized curation of shopping experience for its members?

IE: We use orders, click-stream, and marketing data to optimize offerings, presentation, and communication with our members.

AR: Q3.   As a flash sales portal for luxury goods, are there any unique Analytics challenges faced by Gilt, compared to other online retailers?
Analytics Challenges
IE: The biggest challenge is that every day we create new sales using products we never sold before. There is no previous sales history for these products to use, for example, when making pricing decisions. Most sales are over in 36 hours so the window of opportunity to learn and adjust is very short.

More about it here.

AR: Q4.   What are the key metrics that help you understand consumer behavior?

IE: We look at conversion rate, click-through rate, spent over time, revisit, activation, and retention rates and other typical metrics. These metrics are analyzed by customer segments, and cohorts to better understand underlying drivers. We also calculate brand and category affinities, cart abandonment and return rate, etc.

AR: Q5.   Do you observe any significant change in the consumer behavior across Web and Mobile?

IE: We definitely noticed changes in behavior patterns. Some platforms are used more during the daytime, some more over nights and weekends. Web and mobile users have distinct preferences and respond in a different on communications channels.

AR: Q6. What are some of the most unexpected consumer insights you have discovered so far, through Analytics? Heel Size Analytics

IE: We come across unexpected insights very often but can share only a few. For example, recently we looked at average heel size by state. Some states surprised us. For more details click here or look at fashion blogger's amusing take on it here

AR: Q7.   What is the best advice you have got in your career?

IE: When I read 'Liar's Poker' by fellow Salomon Brothers alumnae Michael Lewis it spoke to me. I left Wall Street and never looked back.

AR: Q8.   What key qualities do you look for when interviewing for Data Science related positions on your team?

IE: In the order of importance:

  • Understanding the scientific method
  • Ability to extract and manipulate data
  • Coding skills
  • Machine learning skills



Incerto Taleb

AR: Q9.   On a personal note, what book did you read recently and would strongly recommend?

IE: I re-read Nassim Taleb's "Incerto" recently and would strongly recommend its four books to anyone dealing with data. "Black Swan" is a good start. I think it should be a required reading to anyone dealing with data.

Related: