20400 Stevens Creek Boulevard must be the only address in Cupertino, California that remains untouched by the IT recession. "Even with the economic downturn, nothing's changed for us," claims Chordiant CEO Stephen Kelly.

If Chordiant has, as Kelly says, dodged the worst of the downturn in IT spending, then it will stand alone among its main rivals in the customer relationship management (CRM) sector. From E.piphany to Siebel Systems, CRM vendors suffered sharp and unexpected declines in business in 2001, while Chordiant has continued to grow its revenues.

Kelly attributes the company's resilience so far to three main factors. Unlike many of its rivals, Chordiant has a small number of large contracts. This means that a significant proportion of the revenue it generates is deferred – and today much of its currently declared revenue stems from contracts signed long before CIOs were forced to start economising.

Second, the size of these deals means that approval – and therefore buy-in – is required from the very top of an organisation. Such senior management buy-in is a key to the successful implementation of CRM projects, says Gartner analyst Gareth Herschel. As a result, Chordiant can boast a "pretty good track record" when it pitches to new clients, he says.

The third factor is Chordiant's January 2001 acquisition of marketing automation software vendor Prime Response. At the time, the two companies were about the same size. As a result, "Chordiant should have grown by 75% even if it was just standing still," says AMR Research analyst Kevin Lucas. Yet while rivals saw their revenue growth slung into reverse during 2001, Chordiant continued to accelerate. Revenues grew by 126% to $76 million in 2001, compared to $33.7 million in 2000.

And, even in the fourth quarter, the company still posted revenues up both sequentially and quarter-on-quarter at $23.8 million, up from the $20.5 million achieved in the previous quarter and $11.6 million posted in the same quarter a year earlier. Furthermore, Kelly says that the company will reach profitability by the middle of this year.

Yet the fourth quarter figures also indicate that Chordiant may be facing leaner times. The company's backlog has dipped significantly, an indicator that sales have become harder to close. Kelly attributes this to Chordiant's acquisition of Prime Response, whose software typically requires shorter implementation projects.

But this should also help offset Chordiant's big weakness: its over-reliance on a few big clients. For example, in 2000, Lloyds TSB Bank accounted for $6.4 million – or 19% – of Chordiant's revenues, while another 30% came from services vendor EDS.

For 2002, Chordiant is hoping that the newly released Chordiant 5 will help re-ignite the rapid growth it enjoyed throughout most of 2001. The Chordiant 5 architecture is based on the Java Two Enterprise Edition (J2EE) three-tier client/server environment. It also includes support for XML.

This ought to enable the company to build more scalable systems, as well as speeding up implementation times. However, at the moment, it only supports IBM's WebSphere application server, which may limit its appeal.

AMR's Lucas describes Chordiant 5 as "a masterpiece of its art", but adds that he believes that it may be "almost perfectionism for perfectionism's sake". One advantage is that such an architecture promises greater scalability, which is particularly important for the business-to-consumer CRM projects that are Chordiant's specialism.

Regardless of how elegant the architecture is, customers will still need to be convinced that the new software will deliver tangible business benefits. A return of business confidence may be more influential than technology perfectionism in persuading cost-conscious CEOs to invest in such big-ticket software.