Pharmas come to the table with lots of data, and information about which doctors are writing what prescriptions.

But they are not using the data with 21st century savvy. "This industry has more info about their customers than most industries, but they are doing less with it," says Kelly D. Myers, CEO of Commodicast at 20 Nassau Street.

Commodicast, based in Santa Fe, has taken tools proven successful in the financial industry and re-designed them to work in the pharmaceutical industry. It mines data on the behavior of patients and physicians, detects patterns in their behaviors, and makes forecasts – all to help medical education firms and other pharmaceutical marketing service companies spend their dollars more efficiently.

"We measure behavior to create consumer-centric health care organizations," says Myers. "We use a lot of fancy technology – genetic algorithms, clustering algorithms, a special type of neural networks, and radial basis function networks – a lot of rocket science kind of stuff, but applied in a fairly direct and logical way."

Myers’ firm will change its name from one with financial overtones to one more pertinent to the pharmaceutical industry – the announcement is expected soon. In August it opened its Princeton office with Alan G. Reicheg, senior vice president of business development, in charge.

"We needed to have a presence in New Jersey, because pharmaceutical is the fastest growing segment of our business," says Reicheg (pronounced ry-tcheg). "Princeton is perfect, for equal access to New York and Philadelphia, plus it has the marketing service companies, and we are not a threat to these companies."

Reicheg had been an economics major at Rutgers, Class of 1989, and he and his wife have three children under six. He has had senior positions in sales and marketing at Carter-Wallace, MedPointe Pharmaceuticals in Somerset, and Savient Pharmaceuticals in East Brunswick.

Myers, the son of a welder and a pharmaceutical employee, grew up near Austin, Texas, graduated with a biology major from Texas A&M in 1979, and has had management jobs at Upjohn and Pharmacia. Myers commutes between New Mexico and New Jersey and says his clothes "live in Santa Fe." He is married and has two children in college.

For Commodicast he partnered with Roger Jones, a Dartmouth-trained physicist who had worked at the Los Alamos National Laboratory, famous for the birth of large scale simulations.

At Citibank Jones had successfully applied non-linear analytic techniques to problems in consumer banking. Jones had identified $200 million in risk in Citibank’s portfolio -individuals with credit cards at risk of going into default – by looking at what they were buying with their credit cards.

"We saw similarities between what banking had gone through in the previous 10 years to what the pharma industry would be going through," says Myers. He decided to employ the quantitative analytics from the financial service industry to help the pharma companies get insights on their customers.

He and Jones founded the firm in March, 2000, within six days of the stock market peak. Despite the stock decline, it is thriving, says Myers, in part because of its focus on pharmaceutical clients.

The company has had income totaling $15 to $20 million over a six year period, both from investors (friends and family) and revenue. "We doubled our revenues over each year we have been in existence, and we had a 5,000 percent increase in revenue over five years. We won an award for the fastest growing company in New Mexico," says Myers. The company is showing a profit.

Commodicast has three core competencies: data mining, pattern finding (teaching computers to find certain patterns), and forecasting. These methods work well with little information, technically called "sparse and thinly populated" data sets. It partners with companies that provide medical education to help make symposia, websites, and written materials more targeted.

"Humans are creatures of habit," Reicheg says. "We do the same things over and over again, with regard to what kind of corn flakes do we buy, which side of the face do we start shaving. Those habits show up in data, computers can be taught to look for patterns in behavior. If you can find a pattern, you can look for an inference about what will happen next."

When Commodicast determines which doctors to target in a particular campaign, it mines the data, such as how doctors refer patients to one another, and which doctors have privileges at which hospitals. With "pattern finding" it studies how the networks are built and how doctors interact. Then it forecasts which doctors will be the thought leaders for that drug.

"Instead of identifying 100,000 physicians, by measuring how physicians influence each other, we may really need to target only 10,000 physicians," says Myers. "Our tools can do more with less."

Pharma marketers already try to find the influence leaders, and they do it by surveying physicians. "Those small surveys cannot include all communities," says Reicheg. "If the survey did not include Tuscaloosa, it will not be valid for Tuscaloosa. Also the surveys rely on what people SAY they do. But our method measures what actually happens."

Forecasting is particularly important when it comes to "compliance" rates. Some patients take their pills religiously, but the noncompliant ones stop taking their medication until, pretty soon, they have forgotten where the bottle is.

So Commodicast determines which doctors have good or bad compliance rates. Doctors whose patients have bad compliance rates might be targeted to get special tools, to help them educate their patients. "We help companies identify appropriate physicians for targeting purposes," says Reicheg, "we don’t actually craft the marketing materials."

Separately, Commodicast has a subscription service, a proprietary interactive database between products, marketers, competitive products, and what is going on in world of pharmaceuticals," says Myers.

Myers predicts that the pharma industry will be soon be cinching its collective belts, and that "do more with less" will be the new mantra. But he encountered unexpected resistance after 9/11, when most companies looked inward, unwilling to try new strategies for awhile. "Traditional large pharmas are reluctant to adopt innovations in marketing, and our progress has been more of a missionary effort than I had hoped it would be. Everybody wants to be first to be second," says Myers. "But now we are getting a critical mass."

What about competitors? None, says Myers. After umpteen presentations, he can predict that the potential client will say: "I’ve never seen this methodology before and how do we make it work?" He has never had anyone say, "Oh yes, that’s like so and so…"

"No one is using this approach," says Myers.

The firm has 12 employees, one in Princeton, two in Boston, and the remainder in Santa Fe. Reicheg is hiring sales and marketing staffers. The firm has had "significant interest" from potential partners and buyers. "We believe we are very innovative and are not as interested in being part of a conglomerate that would not allow us to innovate," says Reicheg. "We are opportunistic and are trying to grow ourselves. Right now, we are having a ton of fun doing what we are doing."

CommodiCast, 20 Nassau Street, Suite 119, Princeton 08540. 609-921-7979; fax, 609-924-7491. Alan G. Reicheg. Home page: www.commodicast.com

Consumer Direct

Both Commodicast and Patient Marketing Group (see page 41) are involved in direct to consumer educational marketing. The difference between the two, says Commodicast’s Alan Reicheg, is that in Commodicast’s consumer centric model, the patient reaches out for information specifically targeted to him as an individual.

He compares it to the travel industry. Ten years ago almost everyone booked flights by calling a travel agent. Now, in booking flights on the Internet, consumers must choose between a dizzying array of search engines and discount websites. But websites like Travelocity personalize the flow of information with where you travel, where you like to stay, and the kinds vacations you like to take. It’s the same with health information. The technology that tracked spending patterns to predict what customers would default on their credit cards can now monitor the progress of a patient with diabetes.

Commodicast takes the website’s patient level data and does data mining to help the website – and its advertisers – motivate patients to improve their health.

A newly diagnosed diabetic, for instance, needs a different set of facts from the 20-year diabetic. Commodicast wants to work with medical websites to keep the frustration factor down by funneling just the right information to each patient.

As it stands now, diabetes patients can be overwhelmed with the information available on the web. Says Reicheg: "So much information is almost a liability."

Facebook Comments