George Eberstadt from TurnTo Networks – Podcast Transcript

Back to EcommercePodcast.com

Shaun Ryan:  [0:04] Hi, I’m Shaun Ryan from SLI Systems, and this is the E‑Commerce Podcast. Today I’m talking to George Eberstadt from TurnTo Networks. Welcome, George.

George Eberstadt:  [0:13] Nice to be here.

Shaun:  [0:14] Now, a traditional question to get us going. What was the first thing you ever bought online?

George:  [0:20] You know, I knew you were going to ask me that, and I wracked my brain. I cannot remember the first thing I ever bought online. But I will say that I am a huge online shopper. I am not somebody who loves shopping in stores. And from the first moment that e‑commerce came out, the whole idea that you could just sort of think about the things you want, and it would magically appear was just a fabulous notion.

Shaun:  [0:43] It is a thing of beauty, isn’t it?

George:  [0:44] Very true.

Shaun:  [0:45] So what was the most recent thing you bought?

George:  [0:49] The most recent thing I bought was a quite fancy new camera for my wife, and it arrived this afternoon. I think she’s going to be pleased. I’m pretty sure she’s not listening to this podcast.

Shaun:  [1:03] We’ll be sure to publish it after her birthday. So, who did you buy the camera from?

George:  [1:09] Actually I bought it from a customer of ours. I believe in patronizing our customers. In fact, there are even some of our customers where I think I probably spend more with them than they spend with TurnTo.

Shaun:  [1:20] Yes, that’s not necessarily a sustainable business model, is it?

George:  [1:24] No, I should barter. It would work out better.

Shaun:  [1:27] So, can you share who the customer was?

George:  [1:29] Oh, sure. Adorama Camera.

Shaun:  [1:31] OK, great. So how was the buying experience?

George:  [1:34] They are really… they’ve got a fabulous site. In this case, I knew what I wanted, because my wife is a photographer, and I had some pretty good guidance. But they have some of the best photography‑related content anywhere on the web. So for people doing their research, it’s really a content destination as well as a commerce destination.

Shaun:  [1:57] Great, great. So, George, can you give me a little bit of background on yourself? How did you get to be where you are today?

George:  [laughs] [2:05] Yeah. I’ve been doing startups since ’97. I was a management consultant before that. And then in ’97 I was consulting to these big businesses on big business questions, and all around me it seemed like there was much more exciting stuff going on. And so I jumped to join a startup, and that was four startups ago. But the last two were businesses that I founded myself.

Shaun:  [2:35] OK, great, so you’re a serial entrepreneur.

George:  [2:38] Yes, I guess so.

Shaun:  [laughs] [2:39] And what were the other startups you were involved in?

George:  [2:44] The first one was a spinoff from Lotus, groupware for the web. The second one was an Israeli company that had technology for online marketplaces. If you remember when B2B marketplaces were going to be the way all business was done. [3:02] The third one, my cofounder invented the key idea at the MIT Media Lab, was interactive conference badges, wearable computing devices for corporate events and tradeshows. And the most recent one, of course, is TurnTo.

[3:17] But the theme that runs through them…If someone is asking, what do those have in common? There’s a social thread through all of them. Every one of them in some way is about how technology can be used to mediate and enhance human interaction.

Shaun:  [3:34] OK, great. And how long have you been at TurnTo? When did you start that?

George:  [3:39] Yes. I founded TurnTo three‑and‑a‑half years ago.

Shaun:  [3:41] OK, great. Tell us a little bit about TurnTo. What do you guys do?

George:  [3:46] Sure. TurnTo provides social commerce tools to online merchants. We’re in that curious place where our products are used by consumers, regular citizens on the web, but our sale is a B2B sale. We actually have two demanding masters. We have to satisfy all the usability demands that end customers have, and all the enterprise demands that businesses have.

Shaun:  [4:16] It sounds very familiar, actually.

George:  [laughs] [4:17]

Shaun:  [4:19] So, can you sort of draw down a little bit more into these social commerce tools? Can you give us an example?

George:  [4:27] Yes, I’m glad you asked. Social commerce, of course, means a lot of different things these days. We focus on the onsite piece, so tools that stores run on their own storefront. Of course, we do provide aspects of the tools that encourage people to post their interactions out to Facebook and out to Twitter, but the tools aren’t running in that environment. We’re not building a Facebook storefront. [4:53] The three applications that we provide for online merchants, first is a Q&A, question‑and‑answer tool that is at the product level. And the secret sauce there is that, unlike sort of bulletin board Q&A systems where the question gets posted, and then it hangs around until somebody finds it and answers it, and maybe it gets answered probably much later or maybe never? Our secret sauce is integrating with the actual transaction data set of the store, which enables us to take that question and email it to people who actually bought the items that the shopper is asking about.

[5:39] So, you can think of it like a Q&A system that’s running on a messaging platform. But the hard part about the messaging is, who do you send the message to? And so that integration with the order database is the key. It’s the foundation of everything. And the net effect of that is that we get wildly more answers, faster answers, and because all the answers are coming from people who really bought the product, better quality answers than any other approach to e‑commerce Q&A.

[6:15] So that’s our flagship product. I think you were about to ask, just the other two, one is a product we call Social Merchandising, which is also, we call it “The See What Your Friends Bought Here” product. We leverage that same transaction data set in a different way. If the shopper chooses to do a Facebook Connect, or use their Gmail address book, or any other social graph source, we can tell them which of their friends is also a customer of that store and what they bought, obviously subject to opt‑in and privacy constraints.

Shaun:  [6:51] Right.

George:  [6:52] And then the third product is a purchase‑sharing product that runs on the order confirmation page, and encourages people who’ve just made a purchase to add a comment, what we call a checkout comment, in response to the question, “Why did you choose this? It’s a super‑friendly tool, and because it runs right there on the checkout page, we find that across our whole network of stores, five to 10 percent of all the items purchased end up getting a friendly little checkout comment. [7:25] And it’s not the same as a review. You know, you haven’t received the item yet, so it’s not an evaluation. They tend to be things like, “Always wanted one, finally got it,” or “A gift for my mom,” or “The best price I could find anywhere.” You know, pithy little things, but they’re nice little endorsements that then, of course, the merchant can use on their site. And along with the Q&A stuff, especially since Google’s been jiggering their search engine algorithms to favor unique content, these approaches to generating unique content, it’s a really powerful way to get unique content onto the item pages on the store.

Shaun:  [8:03] Yes, it makes sense. And I can see that when someone’s just purchased something, they’ve still got that glow of “Yes! I’ve just found what I wanted,” and that’s when they’re going to be probably most positive about the product and say…I imagine you’re going to get a lot of very positive comments at that stage.

George:  [8:19] Yes, we did a sentiment analysis, and the sentiment analysis was .3 percent negative. I’m saying that you just don’t have anything bad to say on checkout.

Shaun:  [8:28] No, no, unless something’s gone wrong with the payment system, or something like that. But it’s almost always going to be positive. So it makes sense, because I know a lot of retailers worry about negative reviews. These are sometimes unnecessary, but it’s still a worry. And so, that’s quite a good way of just getting a whole bunch of positive comments. They’re not reviews, but that’s positive feedback, anyway. So the retailers are going to like it.

George:  [8:52] Yes, that’s right. And we find the same thing is true, too, on the Q&A. Because of this sort of active outreach model, the sentiment that we get on the answers tends to be more positive than what you typically see with ratings and reviews. [9:07] And we spent a fair bit of time looking at that. It’s hard to know totally for certain, but my theory is that that’s because the sentiment that you get from this model is more representative of the overall customer base of the store. But with ratings and reviews…and I love ratings and reviews. I use them heavily when I shop, everyone does. But there is this tendency for bi‑modal distribution. Two percent of the people that are super‑happy, and 20 percent of the people that are super‑unhappy are the populations that are writing reviews.

Shaun:  [9:44] Yes.

George:  [9:44] You know, you get fives and ones. And you get more people who’ve had a one‑level experience that are actually bothering to write about it than the overall population. And that whole middle population, and if you ask them, they’d say, “Yes, it was great, it was what I wanted, it came on time.” They don’t participate, but it turns out that when you email them a question from a shopper, a real person has a real need for help buying…You know, we get an eight‑percent answer rate on that email.

Shaun:  [10:20] Which is phenomenal, because that’s much higher than what you get from reviews, right?

George:  [10:27] Yes, it’s higher than reviews, and it’s higher than just about any marketing email that gets sent out. And so, we know people are responding well to it, just in terms of their willingness to respond. But again, the answers tend to, I think, reflect the overall experience people are having with the product in the store, and most customers are happy. [10:45] So when we look at a distribution, we see, we did a sentiment analysis on answers. It was about a four percent negative rate, which is, it is not zero, but its way lower than what you see when you look at most reviews.

Shaun:  [10:58] Yes, I know, that resonates with me in a couple of ways. I know if I got an email saying some person, Bob, has a question about a product that I’ve just bought, I’d feel more inclined to answer that question of Bob’s than I would a sort of a solicitation to write a review for a product for a retailer. A retailer is a business, whereas Bob’s got a question about a product I’ve bought recently. So I can see I would be much more likely to respond to that than I would to a request to do a review. So that definitely resonates with me. [11:33] And also, recently I bought a Linksys router. But I was reading all the reviews, and there were so many negative reviews on there, really, it almost changed my mind about buying it. But again, I had to think the unhappy people were more motivated to complain than the happy ones are to praise a product.

George:  [11:53] How’d the router work out for you?

Shaun:  [11:54] It was great [laughs] .

George:  [11:57] Whew!

Shaun:  [11:58] Whew, yes. So no, your products are really interesting. So how’s the business going? What sort of success are your customers seeing with these products?

George:  [12:07] You had two separate questions there. I’ll answer the second one first. We’re seeing really, really good numbers from the stores that are using the system. We’re seeing typically more user content overall is being created through the Q&A tool, and the checkout comment tool than most stores’ ratings and review system. [12:31] The sentiments tend to be positive. When you look at the impact on conversion, we measure, it’s not quite a rigorous A/B approach, but we do measure the conversion rate of people who don’t interact with TurnTo at all while they’re shopping, and compare that to the conversion rate of people who do interact with our tools while they’re shopping.

[12:50] And the worst‑performing sites we have show about 200 percent lift. And the best‑performing sites show about eight or 900 percent lift. And I want to quickly make my usual disclaimer about attribution issues. We know there’s selection bias there, but the lift effect, overall, that is a huge effect.

[13:12] And we have stores where 20 to 40 percent of the total orders are coming from people who interacted with the TurnTo system while they were shopping, so it’s clearly becoming an important part of the shopping process, the research process that customers are going through.

Shaun:  [13:33] Great, I mean, they’re fantastic numbers. So I imagine on the back of that, you’re starting to see some good traction.

George:  [13:40] Yes, we sure are. We launched the Q&A product I think in February, maybe early March. And the uptake on that has been really, really good. A whole lot of stores are using it, our pipeline is very full. And what I’m seeing also… we met at Internet Retailer last month. One thing that I noticed at Internet Retailer that I just hadn’t seen before was, the first time stores are starting to come to us with a requirement for Q&A already developed. Whereas in the past, I think people that have used Q&A, it’s sort of been a tagalong product to the main ratings and reviews products that other vendors offer, and often offer it for free. [14:37] And now for the first time, what we’re seeing is stores saying, “Look, we’ve got a set of requirements, and we’re really looking for the best‑in‑class vendor of Q&A.” And they’re starting to look at Q&A as part of their social commerce suite, rather than just the little brother of their ratings and reviews. And that’s something that we definitely believe in, because…like.

Shaun:  [15:05] Of course, yes, that is an interesting trend. Because I remember blogging about, when we first started seeing reviews on our customers, and we saw how we could integrate them Site Search. I remember blogging about that, and it must be about four or five years ago now where we started seeing that trend. And I haven’t seen the same trend yet after the Q&A, but it looks like that’s a trend that’s starting to happen, from what you’re saying. So that’s really interesting.

George:  [15:29] Yes, no, the usual disclaimers are our forward‑looking statements. It’s not like everyone is coming. Most of the people who are adopting our product, when we started the conversation with them, they’re like, “Oh, we didn’t know that existed.” Its more, “Oh, that’s really cool. Yes, sure, we’ll try that. It’s not hard to integrate. Why not? You give us a three‑month free trial, sure, we’ll try it.” [15:54] But there are at least a handful now that are coming to us and saying, “Yes, you know, we have a defined requirement for Q&A.” And last year, we didn’t hear that from anyone. So I think the trend is in that direction.

Shaun:  [16:06] Right, so maybe that’s a trend that we’re going to see a lot more retailers adopting over the next year. It’s in line with the other trends around social that we’re seeing. What else do you think is going to be different about next year compared to this year?

George:  [16:22] Well, I’m not the first person to notice this, but it’s really interesting how sort of after 10 years of very modest innovation in e‑commerce, all of a sudden, the innovation level is just exploding – new business models, new shopping tools, better shopping experiences. I think the overall pattern was in the late 90s, e‑commerce arrived, and “Oh my gosh, I can place an order online!” And it took way longer than I think a lot of people expected just to take care of all the back‑end blocking and tackling needed to deliver on that initial fulfillment. [17:08] And even in the last few years, when probably the biggest objection we’d hear to adoption of our product was, we’re changing our shopping cart, or we’ve got a whole bunch of order management stuff to work out before we can deal with that social stuff. And now, just finally, a lot of stores really have the back‑end stuff taken care of, and they can start to get innovative on the front end.

[17:36] I think the macro trend is innovation, and what particular form that innovation takes, it’s hard to say. The obvious candidates are mobile and social, and even within social, we provide tools that we think fit into three categories of social. But there are 10 or 15 different social commerce application categories out there, all of which have some level of experimentation going on.

Shaun:  [18:05] While we’re talking about new things around social, I’d be remiss if I didn’t ask you your opinion on Google+ that was recently launched.

George:  [18:14] OK, well, I’ll go out on a limb. I did actually just blog in it, so I’ve got some thoughts formed in my mind. My particular take here is with regard to the Circles feature, but it’s pretty foundational to what they’re doing. [18:32] We did something very similar right when we first started. Three‑and‑a‑half years ago when we started TurnTo, not only did Facebook Connect not exist, Facebook hadn’t even come out with their ill‑fated beacon effort yet. But the whole notion of syndicating the social graph hadn’t come out there.

[18:52] So we had an idea for all these commerce applications, and then we thought, but hey, we’re going to have to figure out some way to access the social graph to build these. So let’s be the social graph hub out there. We’ll suck down social graph data from all the sources, and we’ll make it available to anyone who wants to use it. Obviously, Facebook ate our lunch there. We never even got off the ground. It became very quickly apparent that we were not going to be the social graph provider to the universe.

[19:24] But as we thought about that, the very first thing we thought is, “Hey, if we’re going to do that, we’re going to have to have a groups mechanism”, because people aren’t going to want to share everything with everyone. They’re going to want to share some things with some people, other things with other people. So we built a very functional groups mechanism that looked a lot like what Google+ is providing. And we ripped the entire thing out to the roots.

[laughs]

George:  [19:48] What we found was a very painful experience. [indecipherable 19:56]

[19:50] What we found was, people did not want to have to deal with that level of granularity around permissioning. That people want to manage their sharing and their social graph at the level of the complete network.

[20:10] LinkedIn is where I do my professional stuff, Facebook is where I do my personal stuff, Twitter is a very public‑follow model, so that’s where I do stuff with people I don’t even know. And I know what my network is, and I share different stuff in each environment based on this mental model I have of my network. I can’t handle any more granularities than that.

[20:29] So I am going on record here saying I think that the Circles feature is magnificently well‑executed. I love the user experience. I just don’t think that people want to manage their network at the subnetwork level.

Shaun:  [20:51] Yet, I have seen other people say, “Wow, this is just too hard for me to organize and work out who I want to post to.” And I know when I did my first post on Google+, it was saying, “Which Circle do you want to post it to?” And I was like, “Uhh…”

George:  [21:05] Right, it’s hard to figure it out!

Shaun:  [laughs] [21:07] Yes, so I just made it public. And if I do everything public, then I’m not using the Circles feature, really.

George:  [21:15] Well, right there, I think there may be use cases out there where Circles are fabulous, but I think the vast majority of sharing and consuming is going to be from the public feed.

Shaun:  [21:24] Yeah. George, I am mindful of your time. I think on that note that we should probably look to wrap up the podcast here. But I just want to say I think you’ve got some very interesting products there, and it sounds like you’ve got some great results. And I want to wish you all the best for your company, and I want to thank you very much for your time today.

George:  [21:44] Well, I really enjoyed that. As you can see, I love talking about these things.

Shaun:  [21:47] Yes, thanks very much. And that was George Eberstadt from TurnTo Networks. And this is another E‑commerce Podcast in the bag. Tune in next time.