How many infrequent buyers should return each year? 

A brand manager should have a good, evidence-based, understanding of their customer base, i.e. how many people, what types of people, how often do they buy?  Most brand metrics are presented monthly, quarterly, or annually, and this can give an misleading view of the customer base.  A brand’s more regular buyers are over-represented in these short time periods.  Similarly, the sample for much market research (from brand tracking to focus groups) is often deliberately restricted to buyers who have bought recently, again this over-represents heavier buyers.

Previously (Report 73: The Unbearable Lightness of Buying) we showed that around half of a brand’s consumers buy it less often than once a year.  So many of these buyers do not appear in any one year’s brand penetration.  This means that a brand’s consumer base is always larger than its annual penetration figure suggests.

These very infrequent buyers of the brand are worthy of marketer attention.  Firstly, because they are valuable (collectively they contribute almost half of a brand’s sales).  Secondly, because they are vulnerable.  Your brand has the lowest mental availability among these buyers.  When brands decline, it is the sales contribution of this group that drops the most.

Which raises the question, how to tell if a brand is failing among its lightest buyers?  Marketing efforts that appeal to heavier buyers (e.g. price promotions, consumer competitions, new variants, loyalty programs) may prop up sales and temporarily mask erosion amongst the light half of the customer base.

In this report we showcase a useful brand health metric.  We show how much of this year’s sales should be expected to come from buyers who did not buy the brand last year.  For a “fast moving consumer good” it turns out to be substantial – well over one third of annual purchases.

Long-term and continuous brand buying

We constructed a special sub-panel of 12,407 households from standard UK panel data, households who had reported continuously over the five years from 2010 to 2014. From this we analysed the brand performance metrics for 200 large and small brands, and Private Labels in ten different consumer good categories. 

Our focus was to establish in each year the sales contribution of buyers who’d not been present in each brands’ customer base in the prior year.  We particularly wanted to identify if any law-like patterns existed.  Knowing about these would then help any manager understand their buyer base.

What last year’s non-buyers contribute to this year’s sales

The patterns we found were unexpected both in scale and consistency.  Even steady annual brand performance depends on a continuous inflow of many of the brand’s most infrequent buyers. The column averages in Table 1 reveal just how important their sales contribution turns out to be. 

The data show average sales proportions from “new” buyers for the twenty leading brands in each category, in every year. It turns out that just under four in ten sales made by a typical brand are to “new” buyers – those who did not buy it in the prior year. Table 1 summarises how consistent this is.  And note that most of the brands in these categories have well established customer bases, with (polygamously) loyal consumers.  They are not new brands where most of their sales come from recruiting first time buyers.

Table 1: proportion of an average brand’s sales made to last year’s non-buyers

Each year, about four in ten purchases of a brand are made by households that did not buy the brand the year before! 

Applying the Laws of Growth to dig a bit deeper

The sales importance of these ‘did not buy last year’ buyers is very steady from year to year – but there is more variation between categories.  Any instant coffee brand typically relies less on these sales (25%), while olives and razor blade brands depend on them far more (49% and 58% respectively).  This reflects that coffee is bought much more frequently.

Within each category we then found two regular patterns.  First, a Double Jeopardy relationship exists between brand size and the proportion of sales needed from last year’s non-buyers.  Second, Private Label competitors are, by and large, rather less dependent on their lightest buyers each year.  In tables 2 & 3 we demonstrate these market structure patterns in the UK skin care category.

Table 2: UK top skin care brands

Double Jeopardy influences the importance of last year’s non-buyers to this year’s sales. Small brands depend on them more than bigger brands..

Double Jeopardy: There is a correlation (r = 0.6) between share rank and the importance of sales from prior-year non-buyers.  Nivea, the brand leader is the least dependent at 36% while for the smallest brands, Sanctuary and Neutrogena, the proportion is twice that level.  The differential in penetration (brand size) between smallest and largest was over ten times, so the pattern is symptomatic of the Double Jeopardy Law; that small brands suffer twice in having far fewer buyers than bigger brands, and in attracting a slightly lower average repeat rate. Proportionately more of a small brand’s sales are made to very infrequent (lower loyalty) buyers. 

Private Labels: On the other hand, with the exception of Morrison and Lidl (both very small in this category), Private Label brands like Tesco, Aldi or Boots No.7, are less reliant on sales to “new” buyers than the category average.  This is best explained by the fact that these brands compete in restricted markets (eg a Boots private label can only sell to Boots shoppers, not the whole population), so within their restricted sub-markets they are large brands.  In Table 3 we can see that the PL sales to non-buyers are all at around ten points lower than the average, which is broadly in line with the behavioural loyalty attracted by the brand leader.

Table 3: UK skin care private labels (retailers’ own brands)

Summary: moving on from leaky bucket theory

Our key discovery is that there is a pretty consistent benchmark.  For a typical fast-moving-consumer-goods brand about 40% of its sales each year are expected to come from consumers who did not buy it in the previous year.  For smaller brands that percentage is even higher.  For less frequently purchased categories, the percentage is higher.

Or, put the other way around, for stable brands, next year, this year’s buyers will only be worth about 60% of what they were this year.

That sounds like a very ‘leaky bucket’ indeed (i.e. each year the brand loses and gains 30+%).  But that would be a misleading interpretation.  In the same way that many buyers return (simply because they buy every few years) many buyers temporarily disappear (again, just because they only buy every few years).  The returning buyers have not been “won back”, and they almost certainly did not reject the brand and have now been ‘turned around” (Institute Report 61).  Instead the brand has maintained sufficient mental and physical availability to get into the repertoires of a very large number of category buyers, but in many cases at a rather low level.  The typical brand just has many very infrequent buyers, and most don’t buy it every year.

Many ‘activation’ efforts are wrongly predicated on the idea that they win back buyers – the ‘leaky bucket’ idea suggests that without incentives, past buyers won’t return.  But as we previously pointed out in our report on the leaky bucket theory, in 1974 Andrew Ehrenberg wrote that there is no leaky bucket:

Today, we have that successful theory, but Andrew wasn’t quite right.  The bucket does leak a little bit, although there is almost nothing that can be done about this small level of defection (See East and Hammond, 1996 ; Dawes et al 2020 ) And while he was technically right that no special efforts need to be made to ensure that most brand buyers maintain their existing loyalties, we still need to uphold the level of physical and mental availability that underpins them.  The brand management required here is long-term in outlook – the high numbers of ultra light buyers mean that positive or negative results are usually slow to evolve.  For example, turning off advertising, or turning it in the wrong direction, won’t result in an immediate collapse in recruitment and, therefore, market share (see Institute Report 79).  But it will begin the seeping decline of market-based assets, triggering an inevitable share decline. 

The evidence in this report helps to explain why managers don’t need to worry about “winning back” a massive proportion of their customers each year.  But they do need to appreciate the implications of so much ultra-light buying.  And to worry about maintaining mental and physical availability for so many buyers who seldom think about, let alone buy, the brand.

What should I do to look after my most regular buyers?

A brand’s heaviest buyers are individually valuable: our analyses here and in Report 73 suggest that about 20% of the customer base might be expected to account for 60% of total sales. But few brands we have ever seen have come close to, let alone exceeded, 80% of sales from their best customers, so light buyers, like death and taxes, are always with us. Managers are right to ask how best to maintain 40% or more of brand sales every year, from last year’s non-buyers.

How should I target my most infrequent buyers?

Once upon a time, the fact that 30-50% of this year’s sales come from infrequent buyers (so infrequent that they didn’t buy at all last year) would have been interpreted as evidence for recruitment campaigns that targeted ‘lapsed buyers’.  Today we know that is interpretation is wrong.  We will get our expected amount of ‘return’ buyers so long as we maintain our mass availability (mental and physical).

Your heaviest buyers know where you are – they find and buy you quite regularly, and certainly more than once in any single year.  Another third of your customer base are occasional but they are your buyers: they have bought your brand before, but probably a year or more ago. The remainder (and that may be up to half) are so light that they will buy you just once in five years. Your distinctive brand assets (Romaniuk, 2018) are vital to use in advertising to refresh fading knowledge, and to cue buyers at the shelf.  

1. Make it easy for your most occasional buyers to find and buy you

Your heaviest buyers know where you are – they find and buy you quite regularly, and certainly more than once in any single year.  Another third of your customer base are occasional but they are your buyers: they have bought your brand before, but probably a year or more ago. The remainder (and that may be up to half) are so light that they will buy you just once in five years. Your distinctive brand assets (Romaniuk, 2018) are vital to use in advertising to refresh fading knowledge, and to cue buyers at the shelf.  

2. Stay consistent, be distinctive 

In store (and on screen) almost nothing is worse than not being seen. Shoppers look for familiar cues, and this is extremely important for occasional buyers. The familiar makes shopping efficient. If a brand package is radically redesigned, “refreshed” or “repositioned”, then it becomes unfamiliar to the largest part of its customer base.  Their occasional purchases can then easily go en masse to more familiar alternatives.  Our interpretation of the 2009 Tropicana packaging crisis is not that consumers rejected the new design and new logo – they just no longer noticed it on shelf once its distinctive brand assets had been removed.  

3. Buy the broadest reach media you can afford, and be patient

How easy is it for your lightest buyers to forget you? They buy and use other brands more often than they buy yours, but the challenge is not so much the lightness of their buying; it is the sheer number of light buyers that you need to reach.  This is the rationale focussing on using media that does eventually reach everyone. 

4. Deliver competitive parity

Finally, it is important to bear in mind the long inter purchase intervals when market-testing existing and new products.  When your customers do come back to your brand it is important that their experience is just as, if not more, satisfying than it was before.  Experience raises the bar over time, therefore when evaluating cross-sectional consumer perceptions consider how your buyers may be coming back to your brand after a long interval away, using your competitors’ products.  

Conclusion

A brand’s customer base is usually much larger than dashboards and market research reports suggest.  And these buyers buy far less frequently (and more randomly) than is reported.

Every week only a small proportion of a brand’s customer base buy the category, i.e. could buy the brand.  Brands need to maintain continuous physical availability (not start-stop ‘activations’) otherwise some weeks they will miss out on their fair share of these purchases.

And they need consistently branded advertising to maintain mental availability in their large actual, and potential, buyer base.

1  Following the Tropicana pack redesign sales fell 20%. Once the brand had returned to its original format sales returned to their original levels. (Young & Vicenzo, 2009)