After every new camera launches, Bleicher says, his fellow amateur photographers jump online to post complaints about how awful the new cameras are compared to their predecessors. "You're going to see people coming on with blurry pictures of their dogs," Bleicher says. "It's the same with electronic data capture (EDC). Until you understand it, it's going to seem hard to use."

Bleicher is in a curious position. His company is one of a few leading solutions, with various of its software modules found in most major life science companies. But (like other companies in the industry) his most prominent customers are so publicity-averse that he must fend off upstart competitors even now, six years after starting the company. In an environment in which some customers use his software alongside competitors', and other customers develop their own code, he can hardly rest on his laurels.

Bleicher is perfectly comfortable talking about the company's technology all morning. But he is realistic enough to know that there's more to clinical trials than a user interface and a database. "You just want minimal hassles," he says. "Nobody really wants a machine saying, immediately after you enter the data, 'Did you make a mistake?'"

The real-time nature of EDC's feedback loop, requiring more immediate attention to queries, has the potential to disrupt the standard manner of working in some hospitals. The industry's modus operandi, not to mention human nature, is to let paper CRFs pile up in large stacks and then process them in batches. Sometimes that means pulling all-nighters before a data monitor visits to check up on a site's progress.

So for nurses and other personnel drafted into entering data, EDC is often received warily. Why? Work that could be postponed in a paper-based environment becomes impossible to ignore in an electronic study. "The work involved in using paper is the same as doing it online," Bleicher says. "There is an imbalance between their perception of the work and the amount of work."

To hear Bleicher tell it, pilot-project approaches to EDC might not have worked partly because they were attempted on too small a scale. Says Bleicher: "They don't see the advantages immediately. They may not for six to 10 trials." Ultimately, the software and the process will have to evolve together. "There are process issues at the sites," he says of the centers doing clinical trials. "They need to be educated."

Bleicher does sound irked that the small fry are kept alive by tire-kicking pharma sponsors. "This industry has a funny relationship with vendors," he says. "It keeps them alive with a little oxygen. Even those in Chapter 11, getting delisted, they manage to survive."

Patient by patient: Phase Forward's Web-based view of data from a clinical trial.

—Mark D. Uehling