Survey says: success
Faced with pharma resistance, Ivans believed it necessary to develop a different approach, which would ultimately prove more
effective. Instead of matching patients to a prescription database, Ivans started working with comScore, a major Internet
analytics company. He developed a survey-based methodology, where patients exposed to a specific marketing tactic, such as
visiting a website, would be asked questions to determine whether they were a qualified prospect for a particular condition,
whether they had been exposed to a particular advertisement, and whether that exposure had led them to ask their doctor about
the product, get a script, and actually fill the script. Admittedly, the survey approach relies on patients' participation
in the survey, and on accurate self-reporting of their behavior. But Evolution Road clients found this methodology to be as
statistically valid as the PLA approach, while being much faster and cheaper to implement.
The Evolution Road methodologies (both PLA and survey) began to be used by many companies and agencies, including my own,
and it revealed some pretty startling results. While virtually no pharma company is willing to share results publicly, we
consistently saw campaigns achieve ROIs ranging from 4:1 on the low end, up into the mid-teens.
At the same time, using similar survey techniques, our clients also saw ROI for traditional channels showing results that
ranged from barely positive to decidedly negative. Traditional DTC techniques seemed to work well only for new brands with
large budgets that needed to build rapid awareness. Haren Ghosh, Chief Analytics Officer at Symphony Advanced Media, is another
leading player in the advertising effectiveness measurement field, with a long history wrestling with the same problems encountered
by Paul Ivans. "The strength of TV is building brand awareness and influencing people to do research online about the brand,
said Ghosh. "But in lower funnel metrics (i.e., moving past awareness to consideration and trial), online will have higher
ROI than TV."
The story was compelling: except for large brands looking to build mass awareness at launch, the ROI of digital techniques
was clearly superior (and even with large launch brands, digital was proving to be a very effective complement to TV). Yet
budget decisions didn't seem to be much affected by this pretty indisputable evidence. According to Ghosh, year after year,
up to the present day in fact, TV consumed some 70 to 90 percent of DTC budgets. In general, online spend accounts for a single
digit sliver of those budgets.
So why would rational marketers looking to maximize returns for their brands refuse to change their investment behavior in
the face of this evidence?
To get to the bottom of what turned out to be the central question—not whether digital returned higher ROI, but why people
seemed reluctant to believe it—I turned to other veterans of the ROI wars. Hans Sjoquist, president of Global Channel Marketing
Solutions, has spent 32 years in the pharma industry, including stints at Pharmacia, Novartis, and Sanofi. In fact, he worked
on Rogaine, that DTC pioneer back in the 1990s. Sjoquist, like Paul Ivans, was also trained as a scientist, and had been likewise
frustrated by the lack of sophisticated measurement for promotional spend. "No one could ever tell me what the TV dollars
gave me back. If you spent $50 million sales went up, but no one could say exactly why...I saw the Internet as an opportunity
to become more metrics-oriented, more intelligent in terms of how we reach out to our customers."
Back in the late 1990s and early 2000s, "ROI was not a big thing for pharma marketers," recalled Sjoquist. "Most marketers
were measured on launching things on time and on budget. And budgets were based on 10 percent plus or minus last year's budget.
Not very scientific."
Sjoquist was a hands-on client using many of the new methodologies, including work done by Ivans and Ghosh. "When Paul Ivans
came up with proof that if you got more than three page views on a website you saw a higher conversion rate," recalls Sjoquist,
"that was the first attempt to optimize based on a more stringent Rx change metric."