The question sometimes arises whether law-like patterns in buyer behaviour and brand performance can occur in certain sorts of markets. For example, in B2B markets where buyers are thought of as highly informed and rational, one might think multi-brand loyalty and double jeopardy will not be seen. We investigate a market that can be classified as B2B – General Practitioner (G.P.) prescribing drugs to patients. Firstly, let us contrast the typical consumer market to that of G.P. prescribing.

Everyday Consumers
The typical supermarket shopper buys for fewer than half-a-dozen family members. They patronise a small portfolio of retail stores and depend upon the items they wish to purchase being in stock (both for in-store and on-line purchases); if their brand is not in stock, they tend not to leave and go to another store but choose an alternative brand (purchase availability matters). Manufacturers rely on high reach advertising, using a wide variety of media, largely delivering fleeting exposures (mental availability is important). Consumers are thought to care about price and may be tempted by lower prices or promotions even without recessionary pressures. They have experience of many categories and brands within categories and so have much informal expertise, allowing them to navigate stores and make choices very quickly. While brands differ due to distinctive assets, consumers are familiar with the similarities between competing offerings and so may alter their brand choices in seemingly random ways. Despite welcome changes in the sociology of the family, the majority of groceries are still purchased by women.

Physicians
G.Ps. are professionals who receive extensive and continuing training and education. Despite the increasing number of women in practice, the majority of GPs are men. Most GPs in the market we examine work in practices with multiple partners, and while the number of G.Ps. in a practice is similar to the number of people in a household, they all prescribe, whereas one person in a household is usually responsible for most of the grocery purchases. The training that G.Ps. receive should result in treatment choices that are based upon a rational analysis of what is best for the patient in front of them. They can access peer reviewed research which informs the best solutions for specific situations and where there are genuinely superior pharmaceuticals those are the ones which should be prescribed irrespective of cost. G.Ps. have hundreds (if not thousands) of patients and they do not generally pay directly for the medicines they prescribe. They are also indifferent to purchase availability – if the pharmacy has none of the prescribed medicine in stock, it will be ordered via a short and rapid supply chain and the patient will return when it is available – the option to choose a different medication does not exist for the patient. G.Ps. do not pay and in many countries, neither do most patients (or at least not the full cost) and so price is a less significant issue than for groceries. Pharmaceutical companies do not promote direct to consumers, termed DTC (except in the US and New Zealand where DTC is permitted) but they generally do promote to G.Ps., and they have direct contact possibilities via a sales force. Thus in many ways Pharmaceuticals is a B2B market, though patients are the end user of the product. Unlike consumer brands, pharmaceuticals are protected by patents, which means they cannot be copied for the duration of the patent. Table 1 summarises the key differences between GP prescribing and Household purchasing of groceries.

Table 1: Prospective Differences between GP prescribing and Household buying in England

What do we know about consumer markets?

Double Jeopardy
Big brands in any frequently bought category have two advantages over small brands – big brands have more customers than do small brands and they are bought a little more frequently than are small brands. This law-like finding is termed Double Jeopardy.

The number of customers a brand has varies much more than the frequency with which brands are purchased. This means that growing the number of customers matters much more than the frequency with which a brand is purchased – this and Double Jeopardy have been known and discussed for many years (see Institute Report 26: Double Jeopardy Revisited Again and How Brands Grow).

Share of Category requirements (SCR)
On average an individual brand accounts for the minority of household purchases in any category – It is best to think of your buyers as ‘customers of other brands who sometimes buy us’.

Duplication of Purchasing
There is weak/infrequent brand positioning/targeting – we do sometimes see partitions in markets where there are functional differences which matter to consumers, but this is rare. One example is that families with children are more likely to buy pre-sweetened breakfast cereals than are families without children. Aside from this type of obvious functional effect we find no evidence of segmentation. This means that brands should gain customers without specific targeting – see Institute Report 51: Who do You Really Compete With? for a detailed discussion.

What do we see in GP prescribing?

For this analysis we examine General Practice Prescribing in England of a specific type of anti-depressant known as Selective Serotonin Reuptake Inhibitors (SSRis) – There are other drugs for depression, but SSRis are the most widely used and there are 7 significant, different drugs in this sub- category. We picked these products because they are widely prescribed.

We examine key prescribing metrics of all GP practices in England looking at both January 2014 and September 2022 in order to examine patterns over an extended period.

Table 2: Drug performance Metrics – Jan 2014

Readers will note that there are only 6 drugs listed in Table 2 – a seventh SSRi was launched in late 2015 in England, which we will consider later in this report.

The data in Table 2 summarise the actual prescribing of SSRis in the month of January 2014 and it is arranged in descending order of market share (MS). Citalopram was the clear market leader with a share of 30% of SSRi prescriptions. The next column headed penetration (usually referred to as Breadth in Pharma) is the percentage of GP practices who prescribed each of the six drugs and eyeballing the data in this column reveals that the total is over 100% and that is because most GP practices prescribed more than one SSRi in the month. The third column, reports the average frequency with which each of the drugs was prescribed (usually referred to as Depth in Pharma) and the fourth column, computes the average share of SSRi prescriptions satisfied by each drug, The final two columns report the percentage of practices which were 100% loyal to each drug and the average prescription frequency of these loyal practices.

Clearly, there is a very large range of market shares from less than 1% for the smallest SSRi to 30% for the leader Citalopram and we now discuss the other key patterns:

Double Jeopardy
Penetration varies by a factor of 36 between the smallest and largest drug, but prescription frequency varies only slightly (by a factor of about 1.3).

This is the familiar double jeopardy pattern seen in consumer markets – large share drugs are prescribed by many more practices than small share drugs, and the frequency with which large share drugs are prescribed is only slightly more than small share drugs.

Share of Category requirements (SCR)
This measure declines with market share and even the leading drug Citalopram accounts for less than half of its prescribers’ SSRi requirements in a period as short as a month. For the small brands SCR is an even lower percentage.

Thus it is best to think of GPs as ‘prescribers of other drugs who sometimes prescribe our drug’.

In as short a period as a month, less than one third of G.Ps who prescribe a specific drug are completely loyal to that drug, and the ones who are totally loyal prescribe it no more frequently than the G.Ps. who prescribe multiple different drugs. This means that ‘loyal’ prescribers are no more valuable than those who prescribe multiple SSRis – searching for 100% loyalty is not a strategy which will add value, or drive growth.

Duplication of Prescribing
To expand understanding, we now look at some quarterly data, rather than monthly and Table 3 shows the percentage of prescribers of each drug who also prescribed the other drugs in the first quarter of 2014. Looking at the first row of Table 3, of the G.Ps. who prescribed Citalopram, 70% also prescribe Fluoxetine, 68% also prescribe Sertraline and so on.

There is no evidence of partitions in the data. For example, first generation SSRis (Fluoxetine, Sertraline and Paroxetine) might be considered more similar to each other than the second-generation drugs. Indeed Citalopram and Escitalopram are chemically more similar than any other pair of drugs and yet there is no tendency for any pair of drugs to be prescribed more than by virtue of the penetration of any other drug.

Table 3: Duplication of Prescribing – 1st quarter 2014