David Soloff is CEO of Metamarkets, an analytics technology company.
As part of its "State of..." series of articles with industry executives, AdExchanger.com sat down with Soloff to discuss his company, his views on the space, and the state of Metamarkets today.
Click below or scroll down for more:
- Developing Products
- How Predictive Is Media?
- Discovery In Search vs Social
- Milestones Ahead
- The Analytics Shift
DS: Anybody who aspires to build a new technology product needs a set of development partners to do it with and for. That doesn't mean engineers that you hire. It means customers, or prospective customers, who believe enough in your team that they're willing to “throw in” and spend hours on the phone or in meetings with you reviewing what you've built. They're willing to spend precious engineering resources on their side - in our case - to get you the data feeds and figure out how to format that data. Then, they're willing to put their confidence in you such that the product you're delivering back is reliable and customers begin to bet parts of their business on it. At the risk of saying the dreaded "E" word, it's about as close to an enterprise as you can get. The datasets are huge and you have to enter into this iterative development cycle. I don't care who you are, it's going to take months, if not longer, to get something to the point where customers say, "You know what? This really rocks and we're happy to pay you for it."
What is the proposition Metamarkets has made with its clients to get clients to buy into this process?
We deliver visibility into high volume, and high velocity datasets from data-heavy, platform businesses – and, in particular - online advertising which has these coming in from all different directions, and that we capture and store. But making it explorable and comprehensible in any sort of scalable, low‑cost way is a brutal problem from a business intelligence standpoint. That's the “low hanging fruit.” It's a case that's stunningly obvious, but it's also really hard to execute - especially in a low impact way for customers. That's what we've built in the last year. It's very easy for people to integrate with us, especially if they're already on something like Amazon Web Services.
I'll give you some use cases. We have a big, display, real-time bidding (RTB) customer. We also have a big mobile RTB customer. Our product is helping these platform operators, who are both publisher and advertiser-focused, by managing all of the inventory and demand coming on board by seeing who's buying what at what price or how the bidding's going over the course of the day.
If their publisher development or advertiser development teams want to communicate to those different sides of the market on how things are going, they can. Visibility into multiple billions of events every day is not an easy problem to solve - especially distributing that through the web browser or tablet. It's the business intelligence for that marketplace. We can also white label it on behalf of the market. We don't need to be the end provider. It's a very lightweight, flexible infrastructure that provides analytics in a cost‑effective way to that marketplace.
...I'll tell you who I'm thinking of right now... it's the marketer and the person that he's dealing with, the CEO, who doesn't know what we're talking about right now -pretty much. This is a big idea. How predictive is my ad spending going to get? It can be very predictable provided you've also got a model for how people are spending their time. It's all well and good to look at a local dataset like all of the Twitter promoted tweets that people click on, but you need a macro model dictating where is this incremental audience coming from, how are they consuming Twitter, and how is it cannibalizing their more traditional media consumption?
This is the single biggest thing that I see - Twitter and Facebook are fundamentally changing content distribution, not the creation. Yes, the creation of content, insofar as those platforms give everybody the opportunity to say whatever the heck they want to say. It creates a fog, and makes it hard to cut through. Discovery engines - and I know Twitter is focusing a lot on this - discover what is a real signal for you and that's a big problem. That's not the one we're talking about here.
For someone like a traditional content writer, these two platforms are going to dictate who discovers it and maybe even where people consume the media. It's going to be within a new framework. People will either come to it, interact with it or consume it - perhaps all in the same framework. It won't be a broadsheet newspaper or an app that's created and distributed by Comcast, "The Financial Times" or a website. It'll be within the frame of Twitter, a Twitter client, Facebook or some Facebook client that has yet to be built.
Think carefully, though, on how that changes an audience and how, you as a content creator, are going to capture marketing money. If you're a marketer, how are you going to segment your Twitter audience even into people you'd otherwise reach or might overlap elsewhere - but are actually on Twitter because they're reading that article and referring to someone they follow? It's a very complicated, but potentially rewarding, exercise. I don't think anyone's got a concept yet about how to do that. Yes, I do think things can be predicted. I think that it's going to require infrastructure that helps you get a head around all this data.
It's where the impulse originates. With search, it's auto-impulsive. You've got to take the first step. Ultimately, you're the person who has to type things with your fingers, or speak as if you're using Siri, or whatever. As opposed to, if I click a link in someone's Twitter feed, I am raising my hand and I am consuming that content. And, I was in that neighborhood because of who I know, or whom I choose to associate with or follow, even passively, with Twitter.
Twitter's a different model than Facebook so, in that regard, I may never have thought to read an article about "yields" within the topic of "French bonds," I know, at some level, that it's a big deal. But three different people that I follow on Twitter had interesting comments about it, and I did spend 10 minutes reading an article on Bloomberg about it. Not that it's going to do anything for me, or the world necessarily, but I did spend some time, and I think I was even served some ads. That's a real use case. It just happened an hour or two ago to me. So I didn't filter out or look for it. I didn't go searching for an article on French yields. I didn't say, "Hey, what are French yields doing today?" I know now, though.
We've built a wonderful team, and we like each other so I'd like to preserve that at all cost. We're focused and we're not trying to grow as rapidly as possible by just adding people to the team. We want people to be big contributors and be good people. We want to build a solid foundation - that's always on the top of the list because if we don't do that, then we don't have anything. And we've done a wonderful job in the last year and a half and up to almost 20 people. We're still very engineering and product‑heavy, but we've also begun building a business team, which is exciting. We hired an executive from Oracle who ran all of the sales operations for Oracle's sales team. The leverage we've gotten from her in the last couple of months has been insane. We’re now focused on building out the business side of the business and that means account servicing, making sure people are getting what they need, training, getting people into pilots, making sure that they understand what they're going to get in this pilot and how they move from a pilot to a paid relationship. We're building a scalable business, and it's a product business. It's not a consulting business, although, making sure people are comfortable with the product is a critical part of the adoption.
And what about profitability?
What I will say is that if we choose to be, we can be cash‑flow positive. I can say that we're proud to have reached the “land” that many startups never reach, which is revenue. We've been in this new “country” for five months. It’s an interesting, alien land. All of the fruits look like fruits we knew, but they taste different. But seriously, I mean we have revenue. We're pricing nicely with our customers, which we like. Some of them, we do monthly, recurring through a credit card. That's at the lower end. The higher‑end customers, we typically have them in 12 to 18‑month contracts. The base cost, we price according to usage, as in data footprint, which is, roughly speaking, the number of events multiplied by the dimensionality of that data set. So if it's five data feeds, it's a lot more expensive than one data feed – and that's one column. Then, the second column is the number of users supported. It's a simple, straightforward model. And then we have analytics products that we put on top what we charge. We try to be as clean about it as possible. I'm not asking someone to plunk down X thousands of dollars per month for some black-box analytics product. I'm explaining how someone pays for what they get. What I can say about revenue is that we have it and that we will be in the seven‑digits run rate early next year and should we choose to go profitable, or cash‑flow positive, we could do so by the middle of next year. But I don't think we'll do that. I think we'll double down and scale the business.
Our value proposition is compelling for agencies and agency trading desks. How are they managing to buy their activities across multiple services that they're using? How are they applying or understanding how first‑party data drives any lift or performance across those different channels? If they can't merge those datasets and manage them in a cost‑effective way, they don't have a solution. So it's an inverted use case for a publisher that's running multiple ad servers or for a real‑time marketplace that needs to manage buyers and sellers. We've been talking to a number of desks, and actually a couple of DSPs as well so the need for analytics is immensely on the demand side of the market. We happened to start on one side of the market and move towards the middle and if we cultivate relationships on the other, then we thought that'll be great.
Sure. We started out on the publisher side and realized that there's a much bigger analytics hunger in this industry. Simultaneously, the definition of media is changing dramatically. Media now, broadly construed, is any social networking activity. It's consumption of online video. It's even shopping. You've got to ask yourself tougher questions, which are, if you see Twitter with an $8‑billion or $10‑billion valuation in the marketplace, that's premised on something. It's premised on attention and the conversion of user attention and time into dollars, namely via marketing. I fundamentally believe that that the valuation exercise is a zero‑sum game. That $8 billion to $10 billion of valuation has come out of some traditional media company. And by traditional, I mean Yahoo! just as much as I mean Time Warner. That valuation or capitalization, zero‑sum equation, that's a meaningful proxy for attention. Attention is zero‑sum. I've only got so many hours per day unless I'm doing that insane, triple-multitasking thing that I've done and many people do.
Twitter's interesting because they've tapped into the urge to multi-task with the "Mystery Science Theater" TV vamping and people watching TV - and then vamping on Twitter about what they're watching. That's everyone's desire, to have a "Mystery Science Theater" in their own lives where they're riffing off of what they're watching with their friends. It's pretty cool and I think that platform aspires to be more than that. At the end of the day, we are talking about zero‑sum, and we're talking about zero‑sum in terms of competition for marketing dollars. And so I think that just because marketers don't yet understand how to allocate moneys between those different platforms and channels, then therefore, staying out of the market - that's not going to hold.
When I talk about media, I start at the center, the bullseye, with "New York Times," "CNN," "Oprah" and so on. But then I have to begin to move out from that ring, and move into things like Facebook, Twitter, Foursquare and Amazon. I spend more than an hour sometimes reading reviews of records or music on Amazon, and that's in lieu of reading music reviews in "Rolling Stone" or "Spin" as I once did - so I'm very aware of that as well.
And there is the conversion and shopping experience. Shopping is my media. You’ve got to get outside of your head a little bit with that. And it threads back to the Twitter and Facebook ads distribution mechanisms and the discovery they enable. Marketers, just as much as publishers and analytics companies, need to expand their mind. And we're seeing it. It was a big gaming platform that pointed out to us the applicability of our product to that marketplace. They're like, "Oh, I need this." We're like, "Really?" They're like, "Yeah, let me show you how." So we'll see. 2012 should be very interesting. I'm pretty excited for it.
By John Ebbert