This month's issue of Business 2.0 online has an article, "Helping Doctors Go Digital," (first paragraph only for non-subscribers) which describes the success of Epocrates and its medical software for handhelds supposedly achieved entirely by word of mouth recommendations, now popularly referred to as "viral marketing (Wikipedia entry)." They claim 200K physician subscribers, or 1 in 4 U.S. physicians resulting in a 2004 revenue of $23 million.
Early on, Epocrates invited doctors who e-mailed the most effusive praise to become official advocates for the service. Those who agreed received a 20-page PowerPoint presentation to pitch Epocrates to colleagues. Today there are 1,500 such advocates--a formidable unpaid sales force.
However, the business model is to derive the most of their profit from those companies that want to reach their physician base.
Last year Epocrates averaged $25 per doctor in subscription revenue but took in $86 per doctor from ad sales, marketing surveys, and other services that let companies tap into its physician network.
By requiring that their software be updating periodically, "Epocrates can match up doctors' prescribing histories with the ads they've seen and measure response with unprecedented precision. They have expanded their products to include software for diagnosis and treatment reference, med calculations and CME. The article suggests that clinical trial recruitment might be coming in the future.
There is a notable lack of e-prescribing services (something worth investigating). My complaint is the requirement to load their software in the system memory for your handheld, which in my case is a Treo 650. I'd much prefer to load this on my 1 GB SD card to free up system memory which is limited to 23 MB of non-volatile memory. But, this would break their ability to track prescribing histories, AFAIK, but I'll try to confirm this.