Messaging Engagement and Deterioration: Gain a Better Understanding of How Messages Are Working

By L. Andrew Douglas, Senior Vice President

“The advertisements (in a newspaper) are more full of knowledge in respect to what is going on in a state or community than the editorial columns are.” Henry Ward Beecher, Proverbs from Plymouth Pulpit (1887)

In the pharmaceutical world, messaging has been an important part of the marketing and sales strategy through the sales representative. With the advent of digital communication and the ever-changing and less interactive sales environment, it has become increasingly important to quickly identify messages that work from those that do not. Because of competition vying for physician time across communication channels, we can no longer simply rely on the sales rep’s evaluation to identify physicians who are positively responding to communication. Where in the past physician’s acquisition of changing medical information came primarily through the detail interaction and CME, today information can be accessed and questions answered with the simple stroke of a computer key. Standard market research has been ill-equipped to handle the emerging Twitter generation, where poignant, quick and less repetitious information is demanded.

We have developed a good working knowledge of what research methods are best suited in optimizing and testing/tracking message performance. These insights and novel methods can help ascertain how typical message-tracking data can be integrated with secondary data, novel metrics and robust analytics to enable us to get inside the heads of physicians to determine how messages are functioning.

Messages can be categorized into different types of information they attempt to communicate. For example, these can most simply be broken down into three types: descriptive, clinical and experience-based messages. Further, the end criteria for a message can be multidimensional. They can act for market entry, for maintaining product attention, for changing product perception and, finally, for changing behavioral outcomes. We can measure these message types, how they serve different purposes and how they can be successful for disparate reasons.

When doing message testing we use multiple metrics and techniques to identify the optimal message strategy. The reason for using differentiated methods is because messages are “good” for different reasons. For example, clinical messages maybe impactful because they motivate and are relevant; dosing messages may be more differentiated; and experiential messages may be more believable.

“Advertising in the final analysis should be news. If it is not news it is worthless.”
Adolph S. Ochs, The New York Times Magazine, March 9, 1958

The following are some simple thoughts on typical messaging metrics:

Believability of Message

First, believability like differentiation (discussed below) can be bimodal; messages can be believable from an application perspective (i.e., ah ha moments), or they can be believable for already known facts (i.e., the no duh). This can complicate an analysis, thus asking believability in a positive construct or with a bimodal scale can help its interpretation. Furthermore, believability is a great way to show gross negatives in message testing, but really it provides limited variance information across good messages. If the messages have been properly vetted by the marketing and advertising teams, believability can add time with little benefit. However, if the differences in believability relate to a message’s misinterpretation or lack credibility from an experience perspective, that would be a different story and could be tested. Believability from a physician’s perspective is a question of “does the message make sense in the context of what I have seen with my patients?” While patients typically believe what is on “the package,” they believe that if it can be said, it must be true. We typically have better luck with asking physicians about the feasibility of a clinical measure over believability or possibility of attainment of an outcome. Believability is okay if you are testing uncertain endpoints (message development), but less beneficial when discussing optimization and tracking. Attainability of an outcome acts as a better surrogate when evaluating the adoption of a concept.

Differentiation of Message

Differentiation or uniqueness can aid in understanding the drivers of the perception of one product over another. We have found that message differentiation can be bimodal and needs to be asked on a positive construct or on a bimodal scale (a message can benefit from being evaluated as differentiating or it could be a detrimental). However, instead of asking how messages differ on a scale we have found that blinding a message as to the true product ailment and asking message association across market competitors can offer a more robust measure of message uniqueness and is also a more robust measure of message adoption with product alignment.

Relevance of Message

With consumers responding to messages in the CPG space it is common to define relevance as it could refer to differences in interpretation across the population (e.g., the cookie message doesn’t apply to me because I don’t eat cookies, I don’t like this cookie, I don’t like this message-sweetness of the cookie). This is not true in the pharma space as you are cueing a specific ailment or treatment paradigm. Patients and physicians conceptualize messaging and story lines and articulate messages in the context of the corresponding disease state (in reference to their own condition). In fact, patients can typically only talk in the context of their disease state when discussing pharmaceuticals. Therefore, asking about relevance of messages in the context of the disease asks them to define the type of patient they are or what type of patient population the physician treats. A good way to measure relevance is to allow respondents to multi-punch answers or rate messages (not limiting them), allowing for a broad picture of the patient (population). This acts as a good way to define the segments of patients and niche populations.

Persuasiveness of Message

Asking how motivating, or persuasive, is a message is a typical univariate measure of the function of the message. Simply, is the message having a desired impact. However, asking persuasiveness is potentially less reliable than deriving its impact. In addition, with the advent of digital communication, we believe that the function of a message is less to persuade than to engage. When we ask about messaging, we are typically not truly concerned with the recall of a specific tagline only. Rather, whether it be in the consumer, physician or any other research population, we assert that the message as typically discussed is only a product of the marketing team’s need to quickly and efficiently provide information to the recipient. The message itself is a much broader concept and is determined by the engagement of the recipient; it relies on the engagement of its supporting information. Therefore, we propose the role of the marketing message is to peak interest and create desire for more: more information, more understanding, more backup, more support and more product.

Life Cycle of Messages

Messaging, like product lines, has life cycles and these life cycles are linked to the product life cycles. We can measure awareness, consideration, adoption, persuasion, core belief, plateau and, finally, abatement of the efficacy of a message. For a message to be impactful it must be engaged in by the physician, and just as a product, it can suffer at any stage throughout the attrition funnel: A message needs to be adopted, a message must be adopted to ramp up, and inasmuch, a message can reach saturation.




By analyzing the score for the given messages on the characteristics and the variation associated with these metrics we can map the messages in a common space to identify the well performing or deteriorating.



As simple recall does not sufficiently describe the complexity of the message strategy, our methods and analytics respond to the need to move beyond traditional message measurements to better understand message dynamics. Messages are evaluated as to their value and engagement with the physician, allowing the marketer a measure of whether messages are losing their power and reaching saturation. With the use of mapping techniques, we show at a glance how various messages are related to one another, which allows you to visually identify whether messages are acting as intended. And through a novel method of time series analyses we determine message life cycle to optimize a message from growth in effectiveness through message deterioration, which strategically allows for better message targeting.

“The people who flood our living-rooms with a smorgasbord of commercial messages about fetid breath, moist underarms and troubled intestines know this: an appropriate time, place and manner to sell a product is any that sells the product.” George F. Will, “Privacy in the Republic of Appetites,” The Pursuit of Happiness and Other Sobering Thought (1978)