Today's guest post comes from James Ellis. Ellis is the Digital Strategist at closerlook inc. and blogs at digital-pharma.tumblr.com. He also needs more activity on his Twitter account (@digital_pharma) if you'd like to tell him he's wrong. No, really. James will be joining us at ePharma Summit West July 17-19, 2012 in San Franscico, California. If you'd like to join James, be sure to register today and mention code XP1756BLOG to save 15% off the current rate!You have a lot of possible and actual targets, and almost as many ways to describe those targets: deciles (based on past prescribing data), deciles (based on predictive modeling tools), engagement scores, adoption path position, geographic area, specialty, practice type, and more.
The problem with many of these descriptions is that they tend to paint a partial picture of your targets. For example, deciling by past prescription data says ”this group used to prescribe a lot, this group used to not prescribe much.” That’s great information if this were a history class.
Conversely, deciling by predictive modeling (so sexy, so hot this year) is so complicated, most of us can only say ”based on an equation so complicated I'd need PhD and a white board the size of a building to explains it, this group should prescribe a lot and this group shouldn't.” It’s less like history class and more like science fiction.
The issue is that while this kind of segmentation tells you who should write and who shouldn't, it doesn't try to tell you why. And that’s important because if you know why, you’ll know what you want them to do, which means you’ll know what kinds of messages to send them to change their behavior.
So, let's call this this the Context Quadrant Metric (all you MBAs should feel very much at home, they rest of you, don't worry -- this is easy). It's a two-by-two chart. Across the chart, plot out engagement. You can decide how you want to measure this. It can be as simple as decile, but you could also count all the different touches you've made back and forth (number of emails sent, videos viewed, speaker programs completed, honoraria received, etc). Low engagement to the left, high engagement to the right.
Now, you’ve got four quadrants. The bottom-left is filled with targets who are low-engaged, writing few prescriptions. Call these your “Unlikelies.” In the bottom-right you have highly-engaged, low prescribers. Call these your “Underachievers.” Upper-right corner is your highly-engaged, high prescribers. These are your “High-Responders.” Finally, in the upper-left, you have low-engaged, high prescribers. These are your “Overacheivers.”
What does this tell us? Well, we now see where there are correlations between the marketing actions you are taking and prescribing. We’ll also see how in many instances the actions you make have nothing to do with the outcomes. For example, you spend a whole lot of money trying to talk to the Underacheivers, but they don’t seem to be doing what you want. How much money is too much? Or maybe it's proof that your marketing message isn't working for them. Maybe these are the moochers who are happy to watch any videos you put in front of them for the free MREI at the end. Maybe no amount of MREI in the world will get them to prescribe.
How valuable is that information?
And what about your OveracheiversYou are spending a lot of money trying to talk to them, but they are ignoring you entirely without negative consequence to your bottom line. They have massive ROI. Maybe instead of trying to get them to opt in, you should just start sending them thank you emails and leaving them alone. How much money would you save that way?
By understanding the contents of each target group, you can make smarter marketing choices, saving huge amounts of money on dead-end or already-satisfied targets that can be used on new messages to those who will actually respond to them. You can frame marketing strategies (and budgets) around better understanding your audiences.