Sifting through info overload

Quarterly software sales for big-name corporate application makers fell by as much as 27 percent in 2001, but one field held its own: business intelligence software, which finds patterns and trends about companies by pulling information from its databases. Most of the top specialists in that area--such as Cognos and Hyperion Solutions--didn't lose nearly as much ground in the recession.

And Business Objects did even better than its peers. Software revenue for the company rose 13 percent, and net income increased 6 percent, the company reported Jan. 29.

Analysts expect similar growth this year, as the Levallois-Perret, France-based company continues its push into analytic software, designed to give in-depth looks at trends in specific parts of a company, such as sales or inventory. It's a move beyond Business Objects' core business of applications for filtering more general data such as overall revenue.

The business intelligence field is growing, analysts say. Now that most large companies have automated portions of their businesses--such as sales, purchasing, human resources and accounting--they need ways to sift through and use all the information being produced by those systems. Business intelligence software and analytic applications are designed to do just that.

Liautaud recently spoke with CNET News.com about his company's expansion.

Q: Business Objects reported a good quarter and a strong year, but even the business intelligence software space, in general, did not falter as other enterprise application areas did last year. Why is that?
A: I think a couple of reasons.

One, in the time of a tough economy, (business intelligence) is something that companies look towards to help them improve the operations.

They know the revenue of their organization is going to slow down or come down, so they need to protect the bottom line, and to do that, they need to reduce the cost structure. So the analytical capability that they have in systems to know where to cut is critical. So I think that's one point.

Another point, which is a point in a time of tough economy, is to hang on to your customers. You need to have a strong understanding of what the customers do. Business intelligence also plays a major role there.

The third point, maybe, is that on average, business intelligence is a "small deal" business. We don't need to have $10 million transactions to make a quarter. Our average deal size is much smaller; it's less than $50,000. And therefore, we have a much less "lumpy" business than many other software vendors (whose contracts often aren't closed until the end of a quarter).

You're trying to expand beyond basic business intelligence to sell "analytics" applications, but you've said that in general, customers don't really know what an analytics application is. How do you define it?
An analytic application is a solution that gives insights (on) specific areas of the company.

As an example, an analytics application could be for a VP of sales, (who might ask), "OK, how can I increase my customer retention?" We could show how with our software they will be able to analyze the changing behavior of customers over time.

What portion of potential business intelligence customers actually use those kinds of applications?
The (business intelligence) market is penetrated maybe at 10, 15 percent. There's still a huge amount of business to do. The analytics business is still at the very beginning. I think you won't saturate in the coming 10 years.

Products used for extranets and wide networks have been among your fastest-growing products lately. Is that a reflection of an increasing corporate installation of extranets and virtual private networks in general, or are you simply pushing those kinds of products harder?
I think it's a combination of the two. I think clearly companies realize they not only need to transact with other companies--like building online transactions to sell or purchase products with their partners--but in order to be efficient, they need to share information with these partners.

And that's the angle that we've been pushing quite a lot with customers: You want to be online, you want to be more efficient, you want to have a tighter relationship with your customers and partners. That starts with very efficient information sharing.

Do you ever expect it to be the major portion of your business?
That's a good question. I don't know if it will go beyond the core business, but today it's about 20 or so percent. It grew at more than 90 percent this year. I think it's going to continue to grow faster than the core business. Over time it will stay a little smaller, but it's going to be close.

One of the things you've said is that one of the most important criteria, maybe the No. 1 criterion, for business intelligence applications is being independent from other software systems. Yet if you talk to other application guys--database vendors, human resource software vendors, whomever--they generally say the No. 1 criterion is applications integrated into a single system.
(Laughs.) Depends on who you're talking to.

Right, so let's play devil's advocate. If I'm an SAP customer and SAP pitches its own query package or data analysis package, why wouldn't I go with SAP, a company with which I'm already comfortable?
Because if you want to do real analytics, you need to get data that comes from other systems than SAP. Most companies that have SAP also have many other applications on the side, many hundreds of applications...They have Siebel for CRM (customer relationship management), they have PeopleSoft for HR (human resources), and they have built their own over the years.

The benefit that (separate business intelligence software) brings to the customer is...that they can apply that on top of anything, and they can get data from one application, aggregate that with another, and then maybe write back into a third application. These data come from many different systems. We're able to do that easily because we're completely neutral.

You recently hired Phil Solk to fill a newly created position, vice president for worldwide professional services. Why create the new post?
One thing we hope to do is to first of all formalize much more our consulting offerings. Right now, they've been acting pretty locally in defining their own offerings. We need to have a more global and more consistent approach to service offerings for our customers. So he's going to do that.

He's also going to make sure that we can offer our services to a customer to help them to reach the final business benefit that they want to do. As opposed to just telling them, "OK, this is the tool, and you figure out how you go about implementing it," we're saying: "You want for us to reduce your customer churn? Then we're going to help you do that. We have a methodology, we have the consulting services, and we have the (business intelligence) software to help you do that."

From what I understand, these products are generally easier to implement than, say, ERP (enterprise resource planning) software. So, how much service is really required?
You're absolutely right; it takes much less. Nevertheless, I would say right now we do very little business in consulting. We do only 8 percent of our revenue in consulting. We think we can do more. It's going to increase, (but) not by a huge margin.

Also, one thing that we want to do is we want to keep a right balance between what we do ourselves and what we give to our consulting partners. And we have plenty of consulting partners, including the largest systems integrators.

Your company's been around for over 10 years now, and business intelligence software has come a pretty long way. What's ahead? How will the product evolve?
There are several directions.

One that I've talked a little about is this concept of (business intelligence) networks, which is really enabling companies to extend their visibility beyond the four walls of their organization to look at how well their business partners and suppliers and customers are doing. So it's taking the concept of extranet to the next level, using the convergence and Web services to enable that vision.

Wouldn't that require a certain level of access to proprietary data from your partners?
Not necessarily. I think the idea is that in a network of companies who operate in an interdependent way, they all have some information that they have in their own system.

The idea is that they can publish information they want in the form of a Web service...So other companies can now take that data, integrate that with data from other companies or from their own information system, and then have a much better understanding of their overall business.

It's not necessarily a need to have proprietary access; it just means the participants into this (business intelligence) network need to use Web services and business intelligence to expose their data.

The other thing is that I think there's still a huge potential in deploying (business intelligence software) within companies (in a way that's) more embedded into the business processes and (brings) more intelligence. We've seen a few companies who have decided to do enterprisewide deployments, but there's still a lot more to do. So what we'll do...through a combination of enhanced products (and) deployment options (is) enable these companies to deploy to a much larger number of users internally.