Why would there be a ‘health-conscious’ consumer?

A common marketing assumption is that products with a degree of differentiation (i.e. has less fat or sugar) will appeal vastly more to people who value these features more than most consumers. For example, it is assumed that vegan products sell almost entirely to vegans, more expensive products sell mostly to the rich and so on.

The traditional textbook view is that such differentiation means that these products do not (indeed, cannot) sell to the entire market (all category buyers), and thus they do not compete with most of the other products that serve the category. This belief underpins targeting efforts, where marketers try hard to focus their marketing spend on a segment of the market.

However, it is entirely possible for a particular product feature to appeal to all or much of the market even if for most people this is only sometimes. This would mean that, in spite of the products’ apparent differentiation, it really needs to target all category buyers.

In this report, we look at who buys low sugar/fat products and how these products compete with each other. Does it make sense for marketers to target a segment of ‘health conscious’ consumers? Or is it better to target everyone’s occasional interest in healthful options?

If a ‘health-conscious’ consumer segment exists to the extent of being of practical importance, we would see substantial deviations from the laws of consumer behaviour:

• Double jeopardy analysis would show excessive loyalty and deficit penetration for healthful options. That is, we would expect healthful products to appear ‘niched’. So, their sales/share would be made up of fewer households buying (lower penetration) at a higher frequency than expected.
• Duplication of purchase analysis would indicate the existence of a healthful product partition. That is, people who buy any healthful product are much more likely to buy other healthful products and much less likely to buy less healthful products.
• User profile analysis would show two distinct user profiles. That is, healthful products having user bases with quite different socio-demographic characteristics to the ‘regular’ category buyer. 

We analysed our markets by product (i.e. variants) and determined the healthfulness of each product using the Nutrient Profiling Model founded by the UK Department of Health (Department of Health, 2011). This model provides an objective classification of healthfulness for food and beverages using each product’s energy content (kJ), saturated fat, total sugar, sodium, fruit, vegetable and nut content, fibre and protein.  Previous research (Sjostrom et al 2014) looked at products that make “light” claims (e.g. low fat, low sugar), it found similar results to those we present here. 

Note: Our analysis is at the household level. So we are referring to ‘health-conscious households’ when reporting on ‘health-conscious consumers.’

Belief 1: Healthful products enjoy more loyalty than less healthful products (due to the devotion of ‘health-conscious’ consumers)

To determine if healthful products enjoy higher levels of loyalty, we used the double jeopardy analysis (see the Institute’s corporate report #20). The analysis derives ‘expected’ levels of loyalty, given a product’s size, to establish a ‘normal’ loyalty benchmark. Such a benchmark helps in spotting deviations from the norm. However, while deviations from the double jeopardy pattern are known to occur, these deviations are rare. When a product has higher than expected levels of loyalty given its size, a smaller group of people are purchasing it more often than expected. These products are hence known as niche (Kahn et al., 1988). With this in mind, if products have higher loyalty levels than expected, a behaviourally ‘health conscious’ consumer exists.

We systematically analysed all products for all categories over a three year period, finding that a majority of all products followed the (expected) double jeopardy pattern. Of the healthful products, on average 40% identified as being niche (i.e. having higher than expected repeat loyalty but from a smaller than expected customer base). In comparison, a similar 25% of less healthful products identified as being niche.

Therefore, our analysis concludes that, albeit these minor exceptions, people are more loyal to products with higher market shares and not because they have a more or less healthful nutritional profile. As such, our findings show minimal support for the idea that healthful products enjoy much higher loyalty than less healthful products (due to the devotion of ‘health-conscious’ consumers). But, would such consumers display loyalty to multiple, yet only healthful products? We test this assumption in the next section.

Myth busted: A majority of healthful (and less healthful) products have levels of loyalty expected for their brand size, regardless of their nutritional profile.

Belief 2: The ‘health-conscious’ consumer primarily buys from a group of healthful products

The second belief we test is whether ‘health-conscious’ consumers primarily buy healthful products and ignore less healthful products. Such an idea would be consistent with public health advice about consistently including more healthful foods in one’s diet.

To determine the validity of this statement, we use the duplication of purchase law. The duplication of purchase law states that products share more of their customers with other products that have higher market shares, and fewer with products that have smaller market shares (see the Institute’s corporate report # 53). When two or more products share a functional difference (in this case healthfulness), they share more customers with each other and fewer customers with the remaining products. Such deviations are known as partitions and are identified using the Partition Sharing Index (PSI).

If a group of products share customers as expected (that is, in line with size) the PSI would be 1.0. If a group of products shared 35% more customers with each other than expected, the PSI would be 1.35, while a PSI of 0.65 would indicate the group of products share 35% fewer customers than expected. When examining the level of sharing between two groups of products (such as healthful and less healthful products) we use the interpartition PSI, where the same rules apply. Managerially significant partitions/interpartitions that warrant further inspection occurs where 20% of their customers with each other than expected (i.e. interpartition PSI 1.20) (Sjostrom et al., 2014) and 20% fewer of with other products (i.e. interparrtition PSI 0.8). 

If a ‘health-conscious’ consumer segment existed, we would expect to see two distinct partitions with PSIs >=1.20: one for healthful product buyers and the other for less healthful product buyers. We would also expect to see an interpartition PSI of <=0.80. We systematically analysed all five categories for all three years.

Table 1: Partition Sharing Indexes of healthful and less healthful products (2008 to 2010)

Table 1 shows that healthful products share customers as expected with less healthful products. The level of sharing indicates that the two groups are not isolated from one another, confirming that they do share customers. Therefore, our duplication of purchase analysis concludes that healthful products and less healthful products all share buyers with each other, roughly in line with size, regardless of their nutritional value. Our results provide minimal support for the idea that the ‘health-conscious’ consumer primarily buys healthful products.

Myth busted: Healthful product buyers also purchase less healthful products, and less healthful products buyers also purchase healthful products, as expected.

Belief 3: ‘Health-conscious’ consumers are different to ‘regular’ consumers 

Finally, we determine if ‘health-conscious’ consumers are different to other consumers. Such a difference would warrant segmentation and targeting of the ‘health-conscious’ consumer segment, so commonly seen in the industry. 

Previous research has found that the user profiles of competing brands seldom differ (see  the Institute’s corporate report #7). We, however, compare the average healthful product user profile against the average less healthful product’s user profile. We used the approach for a range of typical demographic characteristics to compare user profiles, including household size, the number of children, social class, female head of household age, family life cycle and geographic location. From this perspective, if a ‘health-conscious’ consumer segment exists, we would see mean absolute deviations (MAD) larger than 5.0, meaning there is one group of buyers for healthful products, and another for buyers of less healthful products. A MAD of >5.0 would also be significant enough to warrant further managerial inspection.

Table 2: User profiles for all categories – summary MADs for all years

Table 2 shows that all categories and across all three years, we see that all MADs are small (Table 2). This concludes that ‘health-conscious’ consumers do not differ from regular consumers based on any sociodemographic characteristics, not even social class or household income.

Myth busted: Buyers of healthful products look much like buyers of less healthful products.

Summary and implications

Our analysis of UK panel data from 2008 to 2010, containing almost two million purchases across five categories, concludes that consumers have repertoires of products that span across both healthful and less healthful options within a category. Healthful and less healthful products don’t behave all that differently from each other, with loyalty toward healthful and less healthful products fitting in line with what is expected by the laws of consumer behaviour:

• Double jeopardy analysis demonstrates that excessive loyalty to a single, healthful product (but not less healthful) seldom exists: healthful and less healthful products exhibit loyalty levels expected for their size, not their nutritional profile.
• Duplication of purchase analysis shows there is only a minor divide between consumers who purchase healthful products and those who don’t.
• User profile analysis finds that the sociodemographic characteristics of healthful product buyers and ‘regular’ product buyers are incredibly similar, meaning they are the same buyers.

We conclude that the behaviourally ‘health-conscious’ consumer is more fantasy than fact. It’s most of us, but only sometimes.

With this in mind, the critical take out from this report is that healthful and less healthful products are all just products. For this reason, our findings do not necessitate marketing healthful and less healthful products differently from each other. Which isn’t to say that heathy options should communicate that they are healthy, just that companies don’t need to create separate or ‘special’ marketing strategies based on the nutritional profiles of their products. The most important thing for any product, healthful or not, is to be as mentally and physically available as possible so that it increases their chance of being bought or at least considered in a buying situation.

In light of these findings, we suggest making the following ‘healthy swaps’ in your marketing strategies to ensure you are maximising the potential of your products to be successful: