PM360 OCTOBER 2010 FEATURE

Post-Campaign Measurements for Improved ROI

What’s worse than a disappointing program? Reinvesting in one that doesn’t produce the expected results. With these four key steps, pharma marketers can precisely evaluate their tactics’ success.

BY BOB DOYLE, SDI VICE PRESIDENT OF CONSUMER INSIGHTS & MARKETING EFFECTIVENESS

TODAY’S BRAND MANAGERS ARE BEING TASKED TO CREATE GROWTH AND SUCCESS IN A challenging environment with crowded markets, generic competition, and limited budgets. Pressure to ensure that spending on marketing programs and advertising is efficient and has the desired effect on patients is high. In order for marketers to refine their strategies, accurate and actionable evaluations of a tactic’s success are essential.

Although the types of programs and mediums for advertising may have changed and evolved over time, the need to understand the effects and level of success of marketing efforts has not. In fact, with today’s struggling economy and tight budgets, the need may be greater than ever to prove a positive return on investment (ROI) for marketing programs. Measuring the success of a program or campaign is quite simple, in principle. After running a program,
1) monitor the prescriptions filled by patients involved in the program;
2) monitor the prescriptions filled by patients not involved in the program; and
3) compare the volumes of prescriptions from the two groups. If the patients who were involved in the program filled more prescriptions than the group not involved, then voilà—the program was successful! However, experienced researchers know there’s more to it than that to gather truly accurate results that provide direction for the continuation, refinement, or abandonment of a program.

DIFFERENT DEGREES OF TRUE
There are a variety of study designs and methodologies used for ROI analysis that range from basic to complex, from “express” to custom. The results ascertained from each provide insights with varying degrees of truth. The goal of any marketing team is to get a precise evaluation of its tactic’s success. While it’s bad to have a program that wasn’t successful, it’s worse to reinvest in a program believed to be successful, only to find that the program did not, in fact, have the desired effect on patients. Through our experiences, we’ve outlined the key steps marketers should follow when measuring and evaluating program success.

1) Use Appropriate Methodology and Data
When trying to understand the actions of patients, study the patients—not just prescriptions or sales. Today, the best practices for measuring outcomes and ROI utilize longitudinal patient-level data. Over the past decade, the width and breadth of this data have provided for a rich understanding of patient behavior over time and across sites, including pharmacies, physician offices, labs, hospitals, and retail stores, as well as patient characteristics, including demographics and psychographics—all while remaining HIPAA compliant and keeping patient identity anonymous.

Using patient-level data, and knowing more about patients, has a number of advantages compared to other techniques, but most important it improves the accuracy of the ROI analysis and provides more actionable results for differentiating between new patients, switching patients, new-to-brand, or new-to-therapy patients, so marketers can adjust their programs appropriately.

2) Create a Closely Matched Control Group
A vital part of any ROI analysis is the control group. The only requisite for members of a control group is that they weren’t exposed to a campaign or program. However, to truly attribute changes in behavior among the patients exposed to a program, all other characteristics of the two groups should match exactly. Otherwise, it could have been outside, unrelated factors that affected behavior. The best way to do this is to plan ahead and create a control group during program implementation. However, often because of ethical or logistical concerns, this isn’t feasible. Control groups can also be created after the fact using patient-level data. Evaluations can be done to create profiles of the patients affected by a program and then match them with other patients in the sample. Patient characteristics commonly considered during the control group matching include age, gender, managed care/payment type, co-payment amount, volume of prescription activity, location, and past market activity. However, many other variables can be included, such as patient ethnicity, wealth, diagnosis, and characteristics about the prescribing physician. The more metrics included in this matching, the more concrete are the results that emerge, with the program being the isolated difference between the two groups.

3) Match Metrics to Program Goals
Depending on the challenges facing a brand, success may be defined differently. But once the control group has been created, the success of the promotional program can be measured easily using the metrics that match campaign goals. While increased sales might be the No. 1 goal for the brand team, it would be unfair to say it’s the only objective. Today’s pharmaceutical manufacturers are looking for ways to improve patient outcomes with a host of program types aimed at increasing patient compliance and adherence to therapy. For established brands, patient loyalty and increased adherence are high priorities. Programs designed to improve adherence should not be measured simply by increases in prescriptions.

Other metrics would be more accurate, including persistency curves, average length of therapy, and lifetime values of patients. Because patient-level data is longitudinal, the ongoing prescription behavior of patients should be tracked to evaluate if the program was able to improve adherence. Other tactics, especially DTC advertising campaigns, are designed to help generate awareness or acquire new business. An increasing volume of prescriptions filled by patients new to therapy, which can only be determined by looking into a patient’s history in the market, is a much better indicator of growth than new prescription (NRx) gains. The traditional NRx metric measures only the number of new prescriptions, which are often for continuation of therapy rather than for patients who are truly new to a drug. If the goal of a campaign is to acquire new patients from other brands/drugs in the market, patient-level data can identify which prescriptions are from patients who switched from another brand, as well as at which brand’s expense the business was gained.

4) Know More About Who Was and Who Wasn’t Responsive
To improve the success of a promotional program, it’s important to know which patient segments responded most positively as well as which need improvement. The patients who were most receptive (i.e., followed through with the desired actions) can be grouped and profiles can be created so that more about their characteristics and behavior can be understood. Likewise, the patients who were not responsive to a program can be grouped and studied for a better understanding of how they manage their medical conditions, even if their behavior falls outside the desired actions. The best way to segment patients is to include as many variables as possible to create the most complete picture of the patient. Combining millions of de-identified patient-level healthcare records with consumer and commercial information sheds light on the drivers of patient filling behavior and provides a more complete picture of the patient healthcare experience. When used for segmentation purposes, this information expands the options by which a marketing strategy may be measured and provides additional direction when deciding how programs need to be refined.

Information about some patients and ailments that historically has been missing, but that is now available, involves their consumer packaged goods (CPG) and over-the-counter (OTC) drug use and purchases. While all patients make CPG and OTC purchases, there are some markets in which these play a more significant role in therapy:

1) Pain
2) GERD
3) Skin care (i.e., acne)
4) Osteoporosis
5) Diabetes
6) “Healthy Heart” (obesity, high cholesterol, etc.)

Understanding how consumers are managing their medical conditions with the aid of CPG, food, and OTC products sheds light on the “why” behind patient behavior and provides a better understanding of “path of therapy,” or compliance and concomitant product use. Information like this—which is one of many examples—can be used either alone or with existing primary market research to help refine a program or campaign.

Having a successful promotional campaign that results in improved patient care is its own reward. But that doesn’t change the fact that pharmaceutical marketers are ultimately accountable for the success of their brand and need to be able to measure a campaign’s ROI. When determining the success of programs, it’s important to include key patient characteristics in the analysis, not only to ensure that the control group is appropriate for comparison, but also to help direct where change is needed in the program and messaging.

Bob Doyle, SDI Vice President of Consumer Insights & Marketing Effectiveness, can be reached at bdoyle@sdihealth.com

What’s New in ROI Measurement?
The Physician as Consumer
Looking at more personal physician attributes like wealth, education, media preferences, interests and activities, and ethnicity provides a more complete view of the physician and helps identify all of the possible factors that influence prescribing behavior. Including these metrics in an ROI analysis helps marketers determine what types of physicians responded best and where improvements need to be made, and provides direction for changing programs to better attract or suit the needs of different physicians.

Patients’ Nonprescription Purchases
Including expanded patient characteristics, such as CPG information, which identifies the myriad packaged goods that consumers buy, including food and OTC purchases, helps marketers understand more aspects of patient behavior. This, in turn, can help them refine marketing messages or strategies for reaching potential patients.

TV Tuning Behavior
Technical advancements make it possible to monitor household TV tuning behavior and combine it with de-identified patient-level data to pinpoint what programs patients are tuning in to and whether or not their prescription-filling behavior changes after being exposed to an ad.