Skip to NavigationSkip to content

Does practice make perfect targeting?

Published on 03/06/05 at 10:28am

Since the advent of brick sales information, numerous analytical techniques have become available to help identify the most valuable customers on which to focus salesforce effort. This has now evolved with the development of practice sales data, which attributes pharmacy purchases to individual surgeries using geospatial apportionment.

At first glance, the granularity of this information would appear useful in directing resources toward areas of high potential. However, in reality, how effective is practice sales data for targeting in the modern pharma market?

The three characteristics of good targets

Clearly, whether a practice based or doctor based approach is used to target, the end result of representative activity is particular individuals receiving a detail. It is therefore important to consider the characteristics that identify a doctor as an effective target for your brand. Potential to prescribe sufficient volume within the relevant therapy area is clearly a key consideration, but there are others. Crucially, the doctor's attitude and external influences can drive brand choice as much as representative detailing.

Overall, the value of the doctor can therefore be defined according to these three key criteria:

  • Potential - The opportunity that the customer has to prescribe a product, often measured by prescriptions or sales within a particular therapy area, but equally could be measured by untreated patient potential where customer education is a key component of the message.
  • Environment- Increasingly in today's pharma market, the doctor does not make decisions independently. The increasing control exerted by external bodies such as PCOs and hospitals, in conjunction with the advent of the GMS contract means that ultimately the doctor's personal opinion may not solely dictate their choice of treatment in all cases.
  • Propensity- A measure of the attitude of the GP, which will determine not only their preference for certain therapy areas but more importantly their specific choice of product, distinguishing them into easy and hard to influence groups. Typically, this will vary greatly from one individual to another and depends on their precise demographics and interests. However, personal propensity is often even more important than the environment.

The relative importance of each of these factors will clearly vary depending on the market, resulting in the ideal targeting solution being very brand and organisation specific.

For example, in less differentiated markets share of voice is the key factor in driving brand growth. However, where there are more distinct attitudes toward different products, factoring in the propensity and environment is essential in targeting the right customers to increase brand use.

Essential criteria for practice targeting

Classifying doctors according to the criteria above does not preclude practice targeting as an effective way of reaching the right customers. Indeed, on an operational level there are clearly advantages afforded through the simplicity of a practice based targeting approach.

For example, the representative activity is more tightly clustered into discrete surgeries, rather than pursuing individual targets at various sites throughout the territory, which could improve call rates and time spent delivering the message, provided that accessibility is not an issue within the practice.

However, if this efficiency is gained at the expense of losing some high value customers, then this becomes self-defeating as higher contact rates will be necessary in order to obtain the same return as lower contact rates on more valuable targets.

Therefore, for practice targeting to represent a strategically effective alternative to individual GP targeting, one or both of the following conditions should be true:

  • The doctors with the right characteristics (assessed by potential, environment and propensity) should be sufficiently clustered into discrete practices to capture the majority of the valuable customers.
  • Prescribing decisions within key, high value practices are driven by one, or a few, key individuals, who drive the prescribing of others and these individuals can be identified.

Attitude and environment drive brand choice

Considering propensity first, the personal attitude of a doctor and their confidence to prescribe within a particular therapy area is more dependent on their experience and knowledge than the location at which they practice.

In addition, key GP prescribing drivers will also vary greatly. For example, the relative strengths of clinical efficacy versus cost will vary from one doctor to the next, as will the impact of promotional material identifying the particular advantages of one brand over another. Therefore, not only will the natural prescribing choice vary for each GP, but also their response to salesforce activity, even where the promotional material and representative ability are constant.

This variability in brand acceptance is often apparent in detail follow-ups, where the same message from the same representative is often received quite differently across doctors. It is also illustrated by the wide dispersion across surgeries of doctors who show preference for a particular brand proposition.

IMS identified, through advanced attitudinal profiling, approximately 6,000 GPs with a high propensity for brand X and their distribution by practice examined. Only 10% of practices contained more than one target individual with the right propensity.

Regarding the environment axis, it is natural to assume that the directional influence of the environment will be the same for all GPs within a practice as these factors are largely defined by geography (eg, PCO and local specialist influence).

Indeed, in numerous therapy areas, the overall increasing importance of external influences is clearly visible on GP prescribing. For example, the PCO formulary status for a product can dramatically affect both the overall primary care market share and the detail effectiveness, particularly in the more sophisticated PCOs, which exert a stronger influence on prescribing.

IMS' Xponent Sales Strategy Evaluator (Health Direction Sophistication Index) measured the impact of PCO formulary status and sophistication on brand prescription market share in primary care. In the example considered, formulary status only appears of significant importance in highly sophisticated PCOs, where the representative impact is decreased when the brand is off formulary.

However, within any given PCO, the extent to which this influence will actually dictate the prescribing of the GP will also depend on individual attitude. The interplay of the doctor's attitude and the environment within which they operate will threfore determine prescribing decisions at a product level.

The result is that, where there may be multiple high market prescribers within a surgery, their product prescribing may differ markedly. Identifying those more likely to respond to a detail for your own brand then becomes a key component of effective promotion. Unfortunately, examining historic product prescribing as a proxy for attitude is often dangerous as it is skewed by levels of own and competitor activity.

Furthermore, for new product concepts (repositioning or new launches), there is no guarantee that high prescribing for any current product will be indicative of high potential for the new brand. Indeed, the concept of 'practice innovation' where certain practices show more rapid uptake of a new product is likely to be more indicative of the external influence than a cohesive attitude from the doctors within and does not therefore indicate that any new product will be received in the same way.

Practice data a proxy for patient potential?

Compared to product prescribing, market prescribing for a GP will tend to be more stable over time. So, can practice data at least be effective in identifying doctors with high potential? Examination of practice data does suggest that a majority of market sales come from a minority of practices.

But does this mean that we can say that the practices showing the greatest sales contain the highest value GPs? In reality, one finds that the higher potential practices simply tend to be the larger practices, and therefore would require proportionately more salesforce effort, assuming the potential is evenly distributed amongst the doctors within. Even where there are high potential large practices driven by only one or a few high value individuals, then we are still dependent on the representative identifying them, a task made harder by practice size.

In order to account for this, one can prioritise practices on sales per GP, by distributing the practice sales evenly to the doctors within. In doing so, again the data suggest that a majority of high potential doctors reside in a minority of practices. However, in this instance, one finds that those practices showing greatest potential will tend to be the smaller practices - the inverse of above.

This is because the impact of a high potential GP in a smaller practice will be to lift the average sales per GP for that practice significantly. In contrast, a high potential GP in a larger practice will have their impact 'watered down' and overall the practice will not appear high potential when assessed in this way. The result is that many larger practices, that may contain several good doctors, may be lost, resulting in an inefficient targeting solution. Even within the smaller practices highlighted in this way, there is a reliance on the representative identifying the key GP within them, which may not be obvious and can consume costly time and effort.

The consequence is that, even if the practice apportionment were perfect, then prioritising practices by overall sales potential, or sales potential per GP, will produce quite different target practices. The resultant targeting solution is poor, as it is skewed toward larger or smaller practices, highly dependent on representative knowledge and even potentially inefficient.

Further IMS analysis (IMS Prescribing Indicator for M1A market) found the correlation between practice ranking by total sales or sales per GP is only good for medium-sized practices. The highest value practices in terms of total sales tend to be larger, while higher value practices in terms of sales per GP tend to be smaller.

So, even to simply consider the potential axis of customer value there are theoretical and practical limitations to using practice data to effectively identify where this potential resides at a doctor level.

Practice decisions

Within some practices, prescribing decisions may be made by one key individual, either through referral of patients within the practice to that individual, or by colleagues following clear recommendations from the key decision-maker. In these circumstances, practice data can become more valuable, as the total value of the practice can be exploited through targeting one customer.

However, even within markets where practice decisions are more common (eg, asthma treatments) it is still unlikely that a single individual is responsible for all treatment choices. Whether patients are referred to one person, be that a doctor or practice nurse, or guidelines are issued, the key individual will themselves be influenced by the opinion of their peers.

Furthermore, even in such markets, there will be a mixture of practices where central decision-making is made, and those where decisions are made by the doctors separately. Most importantly, even if one can discern which practices are making central decisions and attribute a potential value to those practices, the key individual or individuals driving prescribing decisions still need to be identified.

While practice data becomes a more useful component of doctor targeting where central decision-making is in operation, one must be careful to consider the cost of assuming that all practices operate as a unit and, more importantly, the cost of failing to identify the right customer where they do.

Where does practice data fit into pharma strategy?

In summary, this highlights some of the dangers in assuming that practice data can be used as a simple, cost-effective targeting tool in an industry where individual decisions are, for the most part, still a key factor in product prescribing and levels of market prescribing differ significantly from one doctor to the next.

The result is that those doctors on whom a detail is likely to generate the greatest increase in brand sales tend to be diffused across many practices.

That is not to say that practice data has no value in today's pharma market. On the contrary, practice data can be an extremely valuable tool when, like any data source, it is used in an appropriate way.

The data is ideal for identifying the behaviour of geographic clusters of practices. For example, when examining the overall response of practices to changing PCO guidance or new drug information, practice data provides rapid feedback. Practice data then becomes a valuable tool for dynamic market assessment at a sub-national and product level.

Answering the question of why changes are happening or how to influence change, comes back to the individual prescriber, however.

Understanding and interacting with them is essential to effective primary care promotion and can only really be answered by approaches that look at individual customers and their interactions, rather than the buildings they work in.


Mission Statement is a leading portal for the pharmaceutical industry, providing industry professionals with pharma news, pharma events, pharma service company listings and pharma jobs,
Site content is produced by our editorial team exclusively for and our industry newspaper Pharmafocus. Service company profiles and listings are taken from our pharmaceutical industry directory, Pharmafile, and presented in a unique Find and Compare format to ensure the most relevant matches