Market intelligence and the formulation of business strategy
Author: Dr Akmal Bhatti – Director Strategy Consulting
The term market research is outlawed within this organization. It conjures up visions of being confronted by an interviewer in the street or on the telephone where one is asked for opinions on a not particularly interesting topic, which will then yield marginally more interesting results and insight. Within the healthcare sector, when one considers an appraisal of developments in the pharmaceutical, biotechnology and medical device markets, there are no standard set of questions which one can ask participants in these businesses. Similarly, it is foolhardy to assume that the response which physicians will have as to their likelihood of prescribing drugs or utilising devices is universal across this specific community of healthcare provider. In times of continuing cost containment and healthcare reform, the art of market forecasting becomes ever more complex. This is an important point given that market opportunities identified from forecasting exercises may be considered the cornerstone of business strategy formulation.
This article attempts to give an overview of the approach used within HBS Consulting to provide market intelligence insight to our clients and how we believe this information can be used in generating business strategies. A most important point to be reiterated here is that each exercise is considered unique in terms of requirement and output and each market intelligence exercise should commence with a blank piece of paper, with the aims and objectives of the client driving the overall research process. This can be considered as the first step in the process.
Forecast models and data input
The complexity and difficulty in accurate forecast modelling is clearly apparent to all who attempt it. It may well be fair to say that the level of difficulty in forecasting sales revenue and market share is higher in the pharmaceutical, biotech and medical device sectors than it is in many other industries. One of the principal reasons for this may be the increased uncertainty companies are faced with when trying to assess the acceptability and recommendation for use of their products by physicians. Increasing the confidence one has in the forecasting methods employed is obviously achieved to some large degree by feeding the best possible data into the model employed. The following diagram gives an idea of the key data components required for forecasting purposes.
Data input is generated from a mix of secondary and primary sources. It is a firm belief (verified time and time again in practice) that great strides in the development of a working and dynamic forecast model is derived from the efficient use of hard primary data. Primary data refers to the process where telephonic and focus group information with physicians, end users and competitors (as a routine example) is used to derive both market and trend data. The use of this primary data in market forecasting is supported by secondary data but the emphasis placed upon secondary data is much less than that applied to the information derived from primary sources. There are sound reasons for this. It should be pointed out that the proof of the power of accessing primary data comes from the highly valuable nuggets of information that this process unearths compared to secondary information.
Taking just partnership and licensing appraisals as an example we have come across numerous companies identified through conducting primary interviews which otherwise would have “slipped through the net.” These companies have been assessed and recommended to clients as a result of conversations we have had during interviews with opinion leaders, physicians and industry participants. The primary interview process yields more valuable information than the background information we class as secondary data. Primary information allows one to capture the mood of the market as it stands whereas secondary information will at times contain a level of bias introduced by the author or authors which is not always supportive or indicative of the real market climate. Primary data has a greater factual component and less of a fictional component than secondary data, and brings with it a dynamism to the forecasting process which can dictate the direction that a market or strategic analysis can take. Market forecasts derived purely from secondary data, and assessment of business strategy purely on the back of this, could be regarded as complacency. Verification using primary data intelligence gathering techniques is essential. This is not a criticism of secondary market data but more a comment on forecasting methodology.
If we rely upon secondary data as providing a sound and verifiable source of the types of data shown on the graphic illustration, one considers that this data is best used as forming the backbone of historical and current market data. It is imperative to develop a clear understanding of the growth rates, market revenues, market shares and pipelines which are important indicators of past and future potential activity. Incidence and prevalence data is best derived from international bodies and associations which are involved directly in the market sector under study. Our preferred option of tackling the compilation of incidence and prevalence data provided by industry associations for example is to increase the confidence in the data provided (some of it very basic and outdated) through the primary interview process with physicians. This allows access or alert to “grey literature”. We have examples where physicians have alerted us to country incidence and prevalence data which they have seen in the most up to date scientific literature and which they feel, collectively, is much more indicative of the national situation than government derived statistics. In certain cases this data has been collected over a period of time within a fairly large urban community with dedicated and tight reporting. Primary data can also provide an indication of consulting rate for certain diseases which can be important in assessing the potential for a device and its uptake. Physician, patient group and government viewpoints on the number of patients who do not seek treatment and may be likely to do so if they are made aware of promising treatments can also be assessed. All this data is then fed into forecasting models.
Macroeconomic factors and an understanding of them are important for an analysis of their impact on policy trends and factors relevant to health economics. The burden of treatment costs for certain diseases is already well documented and acted upon at the government level. Institutes such as NICE in the UK make what, at times, amount to far-reaching decisions on treatment policy and the inclination of government to pay for treatment. In the devices sector we can point to the NICE assessments which support the use of certain devices.
An example of this is the directive in April 2000 which recommended that the selection of hip prostheses for use in Total Hip Replacement (THR) should have as a benchmark selection of those products which demonstrated a revision rate of less than 10% at 10 years. Further to this was what was regarded as a “reasonable recommendation” that products with a revision rate of 3 years having a “performance consistent with the benchmark of a 10% revision rate at 10 years” could also be considered for selection. This widening of the product selection criteria seemed to fit well with emerging policy on reducing waiting lists in elective surgical procedures, where THR was one of the bottlenecks in increased NHS performance. This underlines the fact that investigation of such directives can allow us to sometimes accurately predict policy trends for the short and medium term, and factoring in of these trends in final forecast predictions.
The aim of an analysis of macroeconomic factors should concentrate upon revealing how policy trends shift for a particular disease state, from cure to prevention. This will naturally place differing degrees of emphasis and impact upon a company’s market share within a particular sector.
Price of treatment and the willingness to absorb these costs within national health expenditures at differing levels is predicted. A company has the ability to investigate whether the level it wishes to price its product is acceptable to the health service and where applicable, how far the desired price veers from the acceptable. The value of placing the correct pricing parameters within the model is wholly understood.
Price assumptions will only really come from primary sources. This is a difficult question to answer without companies having understood the outcomes of negotiations with government bodies. As we discussed earlier companies will look for premium pricing and the views of physicians as to whether such charges can be absorbed by health departments. Their willingness to support prescription is useful to compare against company assumptions on pricing. We have explained briefly on how market growth and individual company sales growth is used to investigate how much they fall in line with model assumptions.
In general it is feasible to assume that prediction of trends form the basis of accurate 3 year forecasts. These assumptions can be fed into deterministic models but more tentative assumptions based on longer time- frames should utilise probabilistic simulations. To portray a more dynamic market assessment it is considered a better strategy to continually forecast and assess a market within a shorter 3 year time- frame, forging strategies within these time-frames, as opposed to implementing a long term strategy built around a long term forecasting model.
Assessing physician practice – predicting physician advocacy
Physician interviews and patient focus groups act as sometimes powerful indicators of the achievable market penetration and success of a new drug candidate. Interviews with physicians outside the confines of the “hard sell” approach adopted by medical representatives raises issues and physician concerns which may in many ways contradict the perceptions companies may have of the market, the products in it and their current and planned position within the market. Imperative within this exercise is of course the furnishing of complete data to physicians in order for them to comment clearly on drug properties and their views on therapeutic benefits.
Information from physicians can provide indications of future market share fluctuations, based on the likelihood of use of newer products against established ones. The adoption of this exercise and the indications it provides on prescribing preference also raises the point that forecast models should be developed for three scenarios: the situation where product sales are anticipated to be high, the situation where sales follow the industry norm in terms of growth….a middle ground assessment and then companies should look at the models which depict below average or poor sales. If nothing else looking at three different models and the data fed into them allows for a mid term assessment of the data available and a fine tuning and more confident appraisal of the market and therefore the approach taken in creating the final forecast. This assessment is extremely valuable in creating the picture of the market as it stands and as it will be perceived to be developed. Supplementing this information in a progressive and logical manner is the analysis of competitor activity and the views that competitor companies themselves have of the historical and forward development of pharmaceutical markets.
Forecasting model complexity shows the number of variables that may need to be considered in order to fully understand a market. The above schematic is not relevant for each project but serves as an example of one of the more extensive project undertakings by HBS Consulting.
The knowledge gained is invaluable in organizing a thought process which allows the generation of a strategy. The strategy recommended should follow a logical path and fit with the implementation capabilities of the organization.
Expertly executed market intelligence exercises can be pivotal to the efficient running of companies in the pharmaceutical, biotech and medical device sectors.