When it comes to sales and marketing, the better your information about and from your customers, the more effective your strategies will be, says Anjan Ghosh, head of sales force effectiveness for Sanofi-Aventis. In order to allocate sales and marketing forces appropriately, it’s necessary to have a depth of insight about customers and their behaviors. How, then, do we obtain the data that we need, and how do we use that data for segmentation and targeting? At the eyeforpharma Sales Force Effectiveness – Asia-Pacific Conference in September of 2007, Ghosh spoke on how best to collect and utilize customer insight.
According to Ghosh, there are five key questions that must be addressed when devising a marketing strategy:
• Are we focusing on the right customer mix?
• Are we focusing on the right product mix?
• Are we effective?
• Are we efficient?
• Are we conducting enough sales activities? (in the right places, at the right times, etc.)
We know that improved targeting has a huge impact on sales force effectiveness (SFE), so the challenges then become evaluating and classifying physicians; collecting information to enable informed targeting decisions, and implementing targeting principles and practices appropriately.
Why is customer insight necessary?
Says Ghosh, there are simply too many variables in the marketplace: new and existing customers, covered and uncovered customers, prescribers and non-prescribers of a particular brand, indications, types of patients, specializations, etc. It is impossible to design an effective marketing strategy that would umbrella them all, and there are limitations on the amount and type of information that pharma companies can obtain.
Additionally, no two customers are alike. Patient flow, disease understanding, usage patterns, experience with a pharma company, etc. – all these feed into how a particular physician makes prescribing decisions. And doctors are in different phases of adoption, from burgeoning awareness and interest, to actual trial and usage with patients, to advocacy of a therapy that has worked in their practice. Each stage of the process is best served by its own targeted marketing strategy. Marketing plans based on what Ghosh calls a “micro-understanding” of customers are thus the most likely to succeed.
Determining potential value
Ghosh asks, “What is the level of sales that can be achieved with a given physician?” As resources and access are limited, best practice dictates that the most valuable physicians – the ones with the most potential for future sales – be given the highest priority.
There are many tools for determining the potential value of a given physician; the question is, at what stage is a sales call most likely to influence prescribing decisions? In order to determine potential value of a physician, we can look at:
1. the total number of patients the doctor has,
2. the total number of patients the doctor has who are suffering from a particular condition or disease,
3. the total number of patients who are already being treated,
4. the total number of patients who are already being treated with a drug in the same therapy class as our target drug, or,
5. the total number of patients already being treated with our brand.
As we move outward from the level of highest brand advantage (5), the sales potential increases. If we know where a particular physician sits on the scale from awareness/interest to adoption/advocacy, we can customize our actions to target that position.
Says Ghosh, better targeting is the “most important lever for improving sales force effectiveness.” Fully three-quarters of the overall potential for improvement in SFE come from accurate targeting. And not all doctors are created equally: top-end physicians are far more valuable than an average prescriber, and therefore the lion’s share of resources should be dedicated to those who have the most value and demonstrate the most responsiveness.
Once we’ve determined whom to ask, the challenge becomes finding ways to gather the right data and use it effectively. Ghosh breaks down the process into four steps. First, it’s necessary to design a useful questionnaire; then the survey must be executed correctly and in a timely manner to ensure the data gathered doesn’t expire before it can be utilized. Once the data is accumulated, it needs to be analyzed and used to generate a marketing strategy. Finally, “field realities” that emerge from implementation of that strategy must guide modifications for the next round.
Asking the right (number of) questions
So, how to design an effective questionnaire? Says Ghosh, “As we think about survey design, it is important to carefully consider if we are asking too many questions.” Information about a doctor typically comes from two sources. Profile information includes details that are observable and collectable without having to interview the doctor. These many include name, title, specialization, age, gender, etc. Survey questions fill in those gaps that are not so easily collected. These details may include things like total and indicated patient volume, brand advocacy, prescribing behaviors, length of practice, etc. As Ghosh points out, survey questions are designed to uncover the potential value of a particular physician, not to collect research on the market generally.
Ghosh suggests that short and precise questions be asked and that the number of questions be limited. Medical, marketing and sales departments may want a wealth of information, but often asking too many questions risks discouraging a doctor from taking the time to complete the questionnaire. “Quantitative and objective questions” are best, says Ghosh. “How many breast cancer patients did you see last month?” for example.
Too-long surveys can result in “survey fatigue” and inaccurate answers; too few could result in missed information or insufficient information to differentiate between doctors. Questions should not be open to a lot of interpretation, either by doctors or by reps, as that can lead to biased or incorrect answers. A well-designed survey should ask around four questions per brand. Ghosh suggests using a “cross-functional team brainstorm” to create the questions. He also recommends piloting any survey to determine its effectiveness and the level of effort required from respondents.
The key points to remember when designing a survey are to create surveys that highlight differences between current and potential customers, avoid interpretation, make reasonable demands on reps to execute and do not strain relationships with physicians.
Ghosh emphasizes the importance of including sales in the process of survey creation and execution, right from the beginning. Their field knowledge and relationships with individual doctors can be an invaluable resource in the construction of an effective questionnaire, and field force buy-in ensures that the sales reps appreciate the value of the surveys and don’t regard them as mere “data collection.”
Best practices: survey execution
Implementing the survey in the field presents its own challenges. How long should it take to complete a survey? How will the data be collected? How should reps approach doctors to get the best responses?
Again, Ghosh recommends leveraging the knowledge and experience of the sales force. Survey data is often best collected in conversation rather than from a static questionnaire, and reps can best gauge when enough time has been spent on information gathering. Reps may also be the best judges of where a doctor is in the progression from interest to adoption to advocacy of a particular brand. Some data can come from health care professionals other than the doctor – nurses or administrators, for example – and reps have the best access to these people. Thorough training of the field force will guarantee the best possible results.
Other implementation issues may require “management and execution support,” says Ghosh. To ensure the broadest coverage, surveys should not be aimed only at current customers or targeted physicians but should also include non-target physicians selected by a determined criteria (like specialty). Control totals checked against the expected physician population will help catch high-value, non-prescribing doctors that could otherwise slip through the cracks.
Strong management support can also help alleviate concerns with data processing. How will data be collected? Who will input the information? Do reps have computers in the field? Limited Internet access may also be an issue, particularly in rural areas. Is there manager involvement in the collection or validation of the information? Standardizing the processes by which data is gathered and fed into the system can avoid loss or incorrect information that confuses the results.
How long should the survey take? According to Ghosh, field execution should not exceed one to two months. If the survey takes longer than two months, it’s likely too complicated in the first place. Overly detailed data collection intrudes on promotional time, and if the survey takes too long, the information gathered may become invalid.
Ghosh suggests four questions to help gauge the quality of survey execution:
1. How comprehensive is the resultant data? (What percentage of the total “physician universe” is accounted for?)
2. Does the data promote differentiation between customers?
3. Is the profiling accurate on an individual basis (and were any high-value physicians missed)?
4. How does the situation look from territory to territory? (This last is especially critical in determining allocation of reps and resources.)
Appropriate analysis
In order to get the most value from the information gathered, Ghosh suggests the creation of tiers to divide the physicians into segments. All the many variables that emerge from the data – prescribing behavior, current position on the brand in question – are weighted by importance and used to differentiate doctors. Segmentation allows for far more accurate targeting and thus higher ROI from marketing strategies. Ghosh believes the segmentation should be done at a national level in order to guarantee validation and control.
Incorporating “field realities”
Because the medical profession is not static, information-gathering activities should also be flexible. As Ghosh points out, “practical realities” are not always captured in the data. Doctors move and retire, competitive situations change, new opinion-leaders emerge, access is reduced or improved, etc. Ghosh suggests refreshing survey information annually to keep track of trends and changes in the physician population, as these may necessitate changes in targeting and marketing strategies.
If, for example, a brand continues to underperform despite highly targeted messaging, it may be that the wrong doctors are being targeted. In that case, it may be necessary to improve survey creation or execution or check to be sure that reps are sticking to their target lists. Refinement may involve increasing, decreasing or retraining the sales force, or altering the marketing plan itself. Where there is lackluster performance, surveys can help determine the cause.
As Ghosh says, “physician targeting is a living, on-going process that never ends.” But fully understanding the processes by which a survey is created and executed and the best ways to utilize the data gathered produces the best results and the best return on investment.
Author: Shannon Perry, journalist, eyeforpharma
