Interview With Sam Yagan, CEO Of ShopRunner

We sit down with the serial entrepreneur to get his thoughts on all things data


Sam Yagan has an entrepreneurial resume that puts many in Silicon Valley to shame. He has spent his career building some of the most successful and prestigious digital companies and brands, co-founded SparkNotes in college, which sold for $30 million to Barnes & Noble just one year later, before moving on to start OkCupid, which sold to IAC for $50 million in 2011. Sam then became the CEO of, leading the group through its IPO in 2015. He is currently serving as CEO of Shop Runner, a members-only online shopping service that offers unlimited, free two-day shipping with no minimum order size, and free shipping on returns across a wide selection of today's most popular retailers.

He is active in the Chicago tech scene and as a venture investor, co‐founding TechStars Chicago, Firestarter Fund, and Corazon Capital and sits on the boards of Shiftgig, SpotHero, and Brilliant. Sam has a BA from Harvard, MBA from Stanford, and has earned recognition as one of Fortune's "40 Under 40," Fast Company's "Most Creative People in Business," and one of Time Magazine’s 100 Most Influential People in the World. 

We sat down with him ahead of his presentation at the Big Data and Analytics for Retail Summit, which takes place in Chicago this June 6-7th.

Can you tell us a bit about ShopRunner?

In a nutshell, we provide Amazon Prime for everyone else. We have millions of members who spend billions of dollars shopping at hundreds of retailers in our network.

How important is data to ShopRunner?

For years, we thought membership was the primary asset of the business. When I took over the business 18 months ago, we shifted our lens and realized that it's not the membership itself that is the core value - the core asset of our business is the data that the membership creates. Membership creates data in a bunch of different ways. In order to be in our network, partners have to integrate us into their websites. This means we have both breadth and depth across 140 retailers. First, we get data on all of the traffic across retailers. Of all of that traffic in the network, about 10% of the sales go to our members, so for that 10% of the sales, we see really, really deep data on each member. We see what they were they purchased across the network and we learn a lot about individuals - we learn what items they keep, what items they have returned, what their share of wallet is across the network, and so forth.

Do you hold you all data or do you share it with your retailer partners?

We are in the process of creating various data products that will allow the retailers to benefit from that data - both the data we get from their particular site but also in aggregate the data that we see across the network.

A lot of retailers aren't necessarily data savvy, can they rely on you for insights or should they be building out their own data team?

I think they should really have their own data team, but what we are is a valuable data partner for them. I've had retail partners say, 'hey can I think of you as my outsourced data science department?' and I think the answer is, ultimately, no - you still need to have this expertise in-house. But I do think we can play a really key role in providing a new data source. Any kind of data science that applies to all our retailers, we will build the functionality that enables them to use that. What I think each retailer needs to specialize in is understanding what specific data they need to be looking at, the decisions they need to make, and devote their data science specifically towards solving their idiosyncratic questions, rather than sort of general e-commerce questions which we are much better positioned to answer.

What do you see is the key to building a data-first company?

The most important thing is that you actually have a data asset - I hear all the time companies say they're data-first, but then you dig in a little bit and you realize, you shouldn't be data first.

It's one thing to say that we are an analytical company, it's one thing to say that we make data-driven decisions, but that doesn't mean you're a data first company. I think every company should make decisions based on data, but that's not enough to be, you know, a business where your primary asset is data. I think it's a fine line - as with anything in business, whatever the hot thing is, everybody says that they are, but then you realize that maybe not everyone should be, or not everyone can be.

Interesting. In your experience, do you see a bit of a skills gap when it comes to data?

Of course. I don't think anyone can hire all the data scientists they want. More than anything I think it's a pipeline problem - we just don't have enough people. The demand for data science has just grown much faster than the supply of data scientists.

Do you see this changing as a lot of the data science MBAs that started a few years ago begin producing more graduates?

If you think about computer science as an analogy, it really didn't come up as a major you know until you know the late 1990s, early 2000s. Even now, 15 years later, we still don't have enough engineering talent. It's not immediately obvious to me that the data labor gap is going to get solved any faster.

Do you think more women could be encouraged into STEM a younger age and would that help?

Yeah, that would help for sure. But not so much in the short term. I have a ten-year-old daughter. If the problem you're trying to solve is 'how do you get a ten-year-old girl to like math', that's great in the long term, but she's not going to enter the labor market for 15 or 20 years. I care a lot about girls in STEM, I just don't think that that's the way to solve the labor problem today.

What about machines doing the tasks? Do you see data science being automated?

Data lends itself to machines, obviously. But I still think you're going to need humans on the cutting edge of driving the field forward. Honestly, no matter how much more work machines can do, there's still going to be work for humans in instrumenting, analyzing, and understanding the results. So yes, technology might make it easier, but we will still need a ton of people.

GDPR is coming in over in Europe and there has been a lot of discussions recently around data transparency. Does it impact you?

It definitely affects us. More than anything though, we try to keep up just so you know what's around the corner and what might be coming to States.

Do you think regulations are a good thing in helping to rebuild trust?

I do think it's good for there to be rules that will ultimately increase trust. There should be a set of standards by which everyone is required to act, I'm not close enough to this specific European regulations still whether they're in themselves you know good ideas or not, but certainly, it shouldn't just be every company making up their own rules.

What new technologies are you looking at in terms of data and e-commerce?

For me it's less about technology and more about new sources of data - how can you track someone in a physical store in the same way you can track them on an e-commerce website, how can you gauge intent? You tell by my tone of voice what my intent is, how do we harness this as data? That's what I'm interested in.

In Japan they're doing a quite a lot around in-store experiences to drive new data. Do you see that coming over here more?

I think so. The question is, what are you going to learn from it? Eventually, you're going to walk into a store and the store is going to recognize you. Amazon Go is doing that to an extent now, and that's going to become ubiquitous. But again, I'm most interested in the new data sources that this technology brings to bear.

Would ShopRunner ever move offline?

I'm more interested in bringing the offline data online, though there are certain areas we are looking to do that. Again though, I'm much more interested in going the other way. Let someone else deal with getting the tech installed and then let me use the data.

You can hear more from Sam, along with other industry leaders in the field, at the Big Data and Analytics for Retail Summit. View the full agenda here.


Read next:

Why We Need Data Visualization To Understand Unstructured Data