Business Signatures CEO Speaks on ‘Intent Processing’

In the late 1990s, Peter Relan was working as a software executive at Oracle when his neighbor, Louis Borders, the founder of Borders Books and Music, offered him a chance to join his new startup.

Relan was intrigued enough to make the leap to the dot-com world, where he helped develop much of the technological infrastructure for, and went on to become chief information officer of Borders’ startup, which eventually became known as Webvan.

As Relan readily acknowledges, Webvan was “one of the most spectacular stories” of the dot-com era, “both good and bad.”

Going Public

Before Webvan crashed in one of the largest bankruptcies among Internet era startups, however, Relan was already working on his next project. That company, now known as Business Signatures, formed in the weeks after Sept. 11 and now, after more than three years in stealth mode, is about to go public. Relan is CEO.

Business Signatures releases version 3.0 of its software this summer — the release comes almost exactly three years after Webvan made its now infamous bankruptcy filing — and promises to do for other corporations for whom the Internet is a key strategic business channel what it’s done for an impressive list of customers picked up while in stealth mode.

Though just rolling out its products to general availability, Business Signatures already counts among its customers the likes of Safeway.com, JP Morgan Chase, Geico, H&R Block, MyFamily.com and Electronic Arts.

Customer Intent

Relan spoke with the E-Commerce Times about how things have changed since Webvan rumbled onto the e-commerce scene, how his CIO experience has made it easier for him to talk to and sell to executives worried about e-commerce today, and about how deciphering a customer’s “intent,” something that happens every day in the real world, might be the Holy Grail of online commerce.

E-Commerce Times: Business Signatures has been operating quietly for some time. Why the long period in stealth mode?

Peter Relan: We launched just after Sept. 11. There was a nuclear freeze on business spending and it wasn’t the best time to roll out a new technology product. So we made a conscious decision and spent the past three years developing and perfecting our solution. Instead, we went to CIOs and I was able to tell them, I’ve been in your position. I feel your pain.

ECT: The Web has come a long way as an e-commerce and strategic channel since Webvan was launched. Is the pain any less now for CIOs of companies when it comes to their e-commerce channels? Is it different?

Relan: I think the pain’s greater, the pain is more. In the first few years, there was a lot of pain around questions like what is the right way to build Web sites, presentation and so on. Now the pain is that the Internet has become a strategic channel, truly part of the operating model for a company and no less strategic and important from a competitive view than a call center or branch or store. It’s not experimentation time any more.

ECT: How is it that you managed to attract some fairly high-profile customers along the way.

Relan: What they all have in common is that these people see the Web now as serious — serious customer service and serious revenue opportunities. For them and for all companies that want to leverage the Internet, this is pretty serious stuff now. We’ve focused on corporations that use the Internet as a strategic channel to reach customers or to generate revenue.

ECT: What does Business Signatures offer to do for them and what can your software do for other businesses now that it’s publicly available?

Relan: It’s all about the idea of customer intent. This is something that we don’t do on the Internet but we do in the real world almost as second nature, almost without thinking about it. If you look at the physical channels, this understanding of the intent of the customer is something that goes on all the time. If you walk into the shoe department at Nordstrom’s, before long, someone is going to come up to you and ask if you need help picking out shoes. It’s baked into the process. Or in a catalog setting, if someone is on the phone ordering a bathrobe and they ask if a dress is available in a certain color, right there they are expressing intent. That can be acted on immediately, and in the real world, that happens all the time and it leads to sales and happier customers.

ECT: But that doesn’t happen online?

Relan: Too often on the Internet, the same intent is displayed and the information about it is not available or seen or acted upon. It becomes part of the daily batch of analytical Web data that gets reviewed later on as companies try to make their Web sites better, but by then it’s too late.

ECT: Can you give an example of how customer intent might work in a Web setting?

Relan: A bank has a credit card business and a customer who is online to pay a bill also clicks on a link that takes him to information about high-value balance transfers. But he never finishes that transaction. Later on, that bank can analyze the issue of abandoned transactions. But what you can do about it in real time is create a signature that says when a person makes an inquiry like that but halts the process halfway through, a message goes to the outbound call center. And maybe an hour later, the person is sent an e-mail or gets a call saying “we noticed you.”

Another example of how this is beneficial is in the inbound call center itself. I’ve spent time in the call center of ING Direct — a Web-only bank — listening in on conversations and the first five minutes of many of these calls are spent with the caller explaining to the customer care rep what he was trying to do online and why it didn’t work out. With Business Signatures, the rep could get that person’s user ID from them and within a few seconds of the start of the call, can see an actual log of their recent online history, so they can see for themselves what went wrong. That can free up several minutes of call time, which obviously saves a lot of money and opens up additional opportunities.

The same thing applies on the negative side of intent. Certain actions in a Web setting may indicate that someone intends to commit fraud. Think of what a bank manager would do if someone walked into his branch wearing a ski mask. It’s clear what that person’s intent is and the manager is going to hit the alarm without waiting.

ECT: What changes are necessary to make this technology work?

Relan: It requires no change to applications, no changes to Web pages or templates. It’s completely non-intrusive. When I was at Webvan, I bought a lot of e-commerce software and we vowed that we would not require our customers to change the way those solutions expected us to in the past. Our “intent processor” discovers intent by looking at existing data, by examining HTTP streams as browsers communicate with applications. It’s essentially a conversation. Just as there is conversation going on all the time in human life, the same is true when people interact with a Web site through their browsers. We’re just using the information that is available there.

ECT: Many e-commerce companies do Web analytics in nearly real-time to refine their Web sites and marketing campaigns and the like. How is this different?

Relan: The main difference is in the process. In the case of Web analytics, you are taking the information and storing it in a database, then you are later querying that data to try to give it shape and have it give you insights back. What we do is query the data first, examine the HTTP streams before they are stored. It’s a change in paradigm, from store and query to query then store. Web analytics is still useful, but the value of it is limited by how much information can be stored and how quickly it can be analyzed.

ECT: The e-commerce channel has matured considerably since Webvan had its run. Where are we in terms of e-commerce reaching its potential to capture revenue and maximize business efficiencies.

Relan: We’re past the beginning, but we’re not close to full maturity. I’d say we’re in the fourth inning to use a baseball analogy. We’ve come quite a ways, but there is still a lot more we can learn.