Search our
entire site
Enter your search
terms below, or visit
our search page

Search case
studies only
Enter your search
terms below:

For the table
of contents and
hyperlinks to
general topics
proceed to toc

Piece Of The E-Sales Action

Pandesic's much-maligned business model may start paying off--for its parent companies and its startup customers

ndustry buzz was strong when Intel and SAP launched electronic-commerce joint venture Pandesic LLC in the summer of 1997. Boasting its parents' deep pockets and technology pedigrees, Pandesic sought a place in the big leagues of E-commerce software.

More than 18 months later, Pandesic is something of an enigma. To many industry observers and analysts, the Sunnyvale, Calif., company is a money-losing also-ran, beset by management turmoil and wrenching changes in channel strategies and who it is selling to. To many of its customers, however, Pandesic is a treasured find, offering the only affordable E-commerce package that includes back-end functionality, thanks to its integration with modules of SAP R/3.

This week, Pandesic is set to unveil a deal with a Web-commerce subsidiary of KLM Royal Dutch Airlines that could establish its business model as a viable alternative for E-commerce ventures.

To Pandesic's president, 20-year Intel veteran Harold Hughes, the company is a growing and almost-profitable startup with huge potential and an initial public offering on the horizon. SAP executives, most recently CEO Hasso Plattner, continue to express confidence that Pandesic will be a successful E-commerce player. Plattner and Intel CEO Craig Barrett are on Pandesic's board, and its parents are not hands-off managers. "We're well-funded," says Hughes, "but by very demanding people."

To date, Pandesic's most satisfied customers are Web sellers with revenue below $50 million--some as tiny as Kosher Grocer Inc., an online retailer whose only full-time employees are its two co-founders. The notion of such a small company using R/3 as its back end may seem ludicrous, but that's precisely the point.

In Pandesic's pricing model, the customer pays an up-front fee of $25,000 to $100,000, then pays 1% to 4% of its E-commerce revenue to Pandesic. For that, Pandesic provides and hosts an E-commerce catalog and transaction-processing application based on Microsoft Site Server, as well as back-end fulfillment functions such as warehouse management, shipping, and real-time inventory updates, based on a dedicated R/3 application. Pandesic calls the package "the Pandesic E-business solution."

"We aren't heavily funded, and this was an affordable way for us to grow with them," says Deborah Alexander, CEO and co-owner of Kosher Grocer. The New York food-seller replaced IBM's Net.Commerce with Pandesic in late 1997, primarily because the company didn't want to build its own back end. "When technology is changing every minute, it didn't make sense for us to put all our capital into that," Alexander says.

Companies such as Kosher Grocer may be small compared with Dell or Amazon.com, but by selling to them, observers say, Pandesic provides SAP with knowledge and experience at the low end of the market. As large ERP deployments slow and SAP looks to move into the small-to-midsize market, that experience becomes more valuable.

"Small companies can actually be a more complex sale, particularly if that's not what you're used to," says Steven Tirone, senior analyst with the SAP advisory program of AMR Research. "I wouldn't be surprised if much of Pandesic's experience is percolating up to SAP."

That's not all that's percolating up to SAP. Two weeks ago, SAP unveiled a browser-like R/3 interface called EnjoySAP that was influenced by input from Pandesic. In fact, Pandesic functions as something of a test environment for SAP (and Intel, to a much lesser extent) to gain valuable knowledge that the ERP leader can leverage for its effort to bring R/3 into the world of Web enablement and Internet commerce.

"In Pandesic, they've got a great test bed, now almost self-funding, at a cost that's barely a rounding error to them," says Tirone of AMR Research.

As a standalone business, however, Pandesic has had a rocky time. It took the company a good 12 months to identify its ideal target customer: the $10 million to $100 million Web seller that's either a startup or a separately managed E-commerce unit of a larger company. A year ago, former SAP executive Bryan Plug quit as Pandesic's president and CEO (replaced by Hughes), in part because of a strategic dispute. And last summer, Pandesic scrapped its reseller strategy--at least temporarily--to sell direct.

SAP confirmed earlier this year that Pandesic ran in the red in 1998, which didn't surprise E-commerce analysts. "The non-impact they've had on the market is almost laughable," says Vern Keenan, analyst with research firm KeenanVision. "In the beginning, they had an extravagant ad campaign that seemed to be geared toward the same levels as SAP and Intel. But high-end customers would never give up a percentage of revenue for technology."

Now, Pandesic claims to have found the right approach. "Things change, and you learn," says Hughes. "Our first assumption was that we'd be an adjunct to big SAP accounts. But legacy applications were never designed for the kind of back-and-forth, real-time updates required by Web commerce. For Web commerce, you have to start from scratch."

Making Money Together
That was the plan of Pandesic user Wild Oats Markets in Boulder, Colo., which started selling natural foods and vitamins on the Web last fall. The $400 million, 67-store chain established a separate IT infrastructure and warehouse for online sales. "Our senior management viewed E-commerce as a not-core competency, a 'dip a toe in the pool' venture, so Pandesic's financial arrangement was perfect for us," says Wild Oats electronic marketing director Jay Robinson. "They don't make money if we don't make money."

Like Kosher Grocer, Wild Oats saw Pandesic's inclusion of back-end supply-chain functions as a big plus. "To get real-time inventory and real-time general ledger updates without investing a lot in our own IT infrastructure was something no one else could offer," says Robinson.

Pandesic's business model is like that of an outsourcer. It provides hosting and infrastructure in return for ongoing fees that will grow with volume. "We become their infrastructure in many ways, and what we put in is hopefully returned many times over," says Hughes. But this percentage-of-revenue pricing model isn't always well-received by the venture capitalists backing those companies. "We've had to do a bit of educating there, convincing them that we actually add to the return," Hughes says. Pandesic recently retained an investment banker to help its startup customers raise additional capital.

Pandesic's much-maligned business model, which the company admits contributed to its slow start, may finally be gathering momentum. "E-commerce service offerings are gaining traction, and Pandesic's applications are architecturally well-suited for that," says Aberdeen Group analyst David Alschuler. "Their assets, with an integrated backbone that uses R/3, are unique for companies that are virtual start-ups."

This week, Pandesic will reveal that the cargo division of KLM Royal Dutch Airlines is using Pandesic for its NewFlowers.com Web-commerce venture, a business-to-business site where florists order fresh Dutch flowers to be delivered within 48 hours. For Pandesic, it's a chance to provide the entire IT infrastructure for a new Web-commerce venture, with huge potential business from a very rich, marquee-name parent.

Buy, Don't Build
In essence, Pandesic has staked its future as an E-commerce provider on the time-honored IT decision of "buy, don't build." For Web-commerce pioneers starting up in the mid '90s, "build" was really the only way to go--there wasn't much out there to buy. But Pandesic believes today's E-commerce world will follow the pattern of ERP and other enterprise applications, where in-house software development armies thrive only at a few holdouts.

"SAP took off after their customers tried and failed to build their own systems two or three times," says Hughes. "There were in-house systems trying to take 300 to 400 orders a day and breaking badly. Packages became the way to go--and that's starting to happen in E-commerce."

Web startups such as DVD Express Inc. in Los Angeles may depend on IT, but that doesn't mean they want--or are able--to invest in IT staffs and infrastructure. "Customizing requires a lot of overhead, and it's hard to find a lot of quality technology people in L.A.," says Susan Daniher, VP of marketing at the online seller of digital video disks. "We pay Pandesic a small fee out of revenue [$16 million last year], but in the long run, we're ahead."

DVD Express built its own Web store application first, but went shopping for a vendor when additional volume forced an upgrade. Pandesic's SAP back end sealed the deal. "We wanted to run our own warehouse so customers could see the status of their orders, and SAP was a proven system," says DVD Express chief technology officer Jason Vagner. "For a company rising up out of nothing like us, having something like that to grab on to was great."

Pandesic now has about 50 customers, which Hughes admits is below the company's goal. He says Pandesic is "almost profitable" and its cash flow is "pretty much on plan," and denies published reports that SAP and Intel expected Pandesic to break even last year. He won't comment on specific sales goals but says, "A $100 million run rate would be just wonderful." And he says it's the intention of the company to go public, next year at the earliest.

Can Pandesic make it over the long haul? The apparent technology transfer (or inspiration) in SAP's EnjoySAP is a good sign that SAP, at least, will continue to invest what it needs to in Pandesic's future.

But some observers say deep-pocketed parents can be a drawback in the world of Internet commerce. Analyst Keenan points to Oracle and Netscape's merged Network Computer Inc. unit for Net-enabled consumer devices as a recent high-profile failure. "Well-funded spin-off startups have the chance to make many mistakes before they go belly-up," he says. "But they seem to lack the fire in the belly of real startups."

Pandesic's Hughes couldn't disagree more. He points out that 140 of Pandesic's 160 employees are from neither SAP nor Intel, giving the company a highly entrepreneurial culture. Its nondescript headquarters, where a security guard can double as receptionist in the middle of the day, certainly fits the mold. Hughes himself, after 20 years in Intel operations, now has Pandesic stock options instead of Intel stock options. He says he's living the same long days and sleepless nights as any young Silicon Valley CEO.

The E-commerce industry is looking for Pandesic to earn more of its money, and rely less on its parents' deep pockets.

TABLE OF CONTENTS