Tracking Customers While Preserving Their Anonymity

Opion CEO David Holtzman says his company can gather vital marketing data from the Net without having to identify who it comes from

By Jane Black

Stark battle lines have been drawn in the debate over online privacy. Consumer advocates hate the idea that companies are collecting personal information about them under the guise of better marketing. Companies claim they won't abuse this data and that the info helps them better serve customers. Congress keeps making noises about legislating on the issue. But there's a big problem: With the sides so far apart, nobody knows where the middle ground is.

Enter Opion, a technology startup whose products scour Internet message boards for valuable marketing information. The company tracks which messages and information are spreading quickly on the Internet. Rather than snag consumer data on individuals, Opion assigns the users it tracks a number that corresponds to a screen name rather than a real name. So Opion can predict consumer trends without violating privacy. And its strategy doesn't rely on building large databases of personal profiles to create marketing information.

What's more, the company claims its tracking info is more predictive of where product trends are going, rather than reactive -- one of the raps against traditional data mining. Opion thinks its tools will allow it to more effectively analyze trends without ticking off consumers.

Recently, I sat down with Opion Chief Executive Officer David Holtzman to talk about this fresh approach to gathering consumer information and whether Opion can both satisfy companies desperate for market research and consumers who want their data protected. Here are edited excerpts from our conversation:

Q: What do your products do?


One of the technologies we created can determine who is influential, and the degree of influence. Basically, we're able to assign a number to an individual, given the context of a subject, or to a community. Based on the degree of influence, we know that when these people say things [on the Internet], they can move things. We now have a database of millions of people on the Internet, pseudonymously. We have no connection with the real name, and don't want any. We never want any.

I would argue this is just an extension of the role of the media in the last 30 years. You look at some of the financial analysts -- the Henry Blodgets and people like that. I would argue that they're just a pseudonym with a picture. Because in that sense, they speak and people listen, and the markets move. The fact that there's continual issues with the SEC involving people posting on the message boards should tell everybody that it's moving things, and changing things.

Q: Will such influencers eventually surpass the media in their roles?


Burston-Marsteller did a big study called "The e-fluentials." They went out with Roper and interviewed 2,500 people whom they thought were influential on the Internet, based on subjective analysis, and then sent them a note and asked them if they could interview them.

They found that almost all of them were professionals. I mean, they weren't geeks sitting in a basement somewhere. They were bankers who were talking about the law, and lawyers who were talking about stocks. And then they tracked them.... Finding out who these influencers are not only gives you a sense of forecasting on where things are going, but it also gives you the ability to promote to people who can really make a difference.

Q: But isn't a company still better off to have actual customer data?


No. If you look at the way most of these companies work right now, with their CRM [customer-relationship management] systems and data warehousing, really what they're doing is extrapolating from their database. They say, "We sold X product to 20,000 people who all looked like this demographically. So, that means everybody who looks like this demographically will probably want one of these, too. And even if only 1% does, and 99% get offended, who cares, because you know, it's 1% more sales." So, if the cost of the marketing campaign is less than the 1%, they're going to make profit out of it.

It's that kind of attitude that's causing the problem. But if they have true forecasting, if they can actually see where the products are going, then they'll realize that the customer database isn't a good road map.

Q: So, a company's motivation for choosing to use this product has nothing to do with users' privacy?


No. It's [purely] self-serving on the company's part. By abstracting out a concept like this, where you have an intermediary layer to hook up consumers with companies, based upon interest, I think it's a win-win. From the company's viewpoint, I don't think they [could] care less about privacy. I would never sell privacy to them. What I would sell is something that's so much better than anything they can get. And if, oh by the way, it's just happens to be privacy-clean, that's great for everybody.

You have to look at both sides. I think businesses have legitimate reasons to try and collect information -- they need to increase revenue because of commitments they've made to investors. And the only thing they have to increase revenue on, typically, is by selling more stuff to the customers that are already there, or their immediate and symmetrically similar friends. But they're going about it the wrong way. By finding a new way, they can get what they need and satisfy consumers' desire for privacy.

Q: Sounds good. But it will take time to integrate such a radical marketing approach. In the meantime, Congress keeps talking about passing some sort of privacy-protection legislation. What's your vision of the best privacy legislation?


I think the best privacy legislation is to find out who's actually doing it, and put real-world sanctions on the individual. Problem is, who do you get? You get the content generator. You got a direct marketer. And then you got some marketing guy in the company, but it might not even be in the company. It might be a joint venture. There are so many variations now, it's not clear you're going to find out who to get. And if America puts up some big, obtrusive roadblocks that they can't enforce, everyone will just say, "Well, the hell with this. I'm going to move everything to Europe. My customers can still hook in from America."

So, the only solution then becomes something extremely draconian, which is you do something like you put monitors or controls down on the personal computer, on the hard drive, or somewhere close to the consumer, under the guise of privacy. That really scares me.

Q: So do you think there shouldn't be any legislation?


No. I think legislation is good, because I think it shows the country is taking an ethical stand. I think it's good for that reason alone. An amendment or something would be nice -- something that creates a concept and definition of privacy.

Privacy has to be cultural. In addition to legislation, I think we need an informed discussion on this subject so that privacy can become an extension of culture, rather than an extension of legislature. If people believe in this and they understand it, and they define privacy as a degree of what it means to give their information out, then it creates an environment where businesses make more money by supporting that. Then I think we have a good world. That's what my company is supposed to be doing.

Q: How long do you think it will take to create this culture of privacy?


I think it's half a generation away. There's a tendency in American rhetoric to find an issue that disturbs people, and draw a line in the middle of that issue, and everybody jumps on one side of the line. I think both sides of this debate tend to be very extreme. It's becoming: Are you for privacy or are you against privacy. Well, I'd like to believe that everybody's for privacy. We just need to look for compromise solutions.

Black covers privacy issues for BusinessWeek Online in New York

Edited by Alex Salkever

Before it's here, it's on the Bloomberg Terminal.