Double Jeeopardy Revisited
Andrew S. C. Ehrenberg, Gerald J. Goodhardt, & T. Patrick Barwise
Double Jeopardy Revisited
In any given time period, a small brand typically has far fewer buyers than a larger brand. In addition, its buyers tend to buy it less often. This pattern is an instance of a widespread phenomenon called
"double jeopardy" (DJ). The authors describe the wide range of empirical evidence for DJ, the theories that account for its occurrence, known exceptions and deviations, and practical implications.
I
N this article, we review a little-known but widely occurring and theoretically supported regularity in competitive markets, namely that small brands generally attract less "loyalty" among their buyers than large brands do among theirs. Twenty-five …show more content…
In allocating marketing effort it is worth knowing that it is normal for a small brand to attract somewhat less
"loyalty" and yet to survive. We return to these issues in the final discussion.
All the general explanations of double jeopardy are statistical, relating to the size structure of the market.
Other things being equal, small brands attract less loyalty just because they are small (i.e., have lower market shares). No other marketing mix or consumer variables need be invoked to explain DJ. McPhee himself outlined this notion in broad terms and it is also common to the later, more detailed models reviewed here.
What, then, is the role of marketing factors such
Journal of Marketing
Vol. 54 (July 1990), 82-91
as product formulation, price, distribution, advertising, promotions, and market segmentation? The answer seems to be that these factors give brands their different sales levels, which in turn show up in the
DJ pattern, but rarely cause big additional differences in brand loyalty. Hence competitive brands tend to differ mainly in how many buyers they have …show more content…
These brands obtain a somewhat higher average purchase frequency (about 20% or more) than their market shares or penetrations would warrant, in comparison with the w(l - b) or Dirichlet models (Ellis
1989). The explanation is not that repeat-buying loyalty for private labels is higher than normal, but that consumers ' access to any particular supermarket chain 's private label is necessarily restricted. In other words, it seems that the penetration b for private labels in the w(l - b) model is low rather than that the repeatbuying measure w is high; within their own stores, private label brands tend to have very high market shares. An occasional deviation seems to occur also for major market leaders. In our consulting experience, some leaders have a somewhat higher-than-expected level of repeat buying (i.e., they behave like an even bigger brand than they already are), but this deviation does not occur universally.
Summary Conclusions
Overall, the message is that frequent-purchase markets almost invariably show a regular or "lawlike" double jeopardy pattern whereby smaller brands tend to attract somewhat less loyalty, with deviations that are mostly relatively small and so far little understood. Not many such empirically and
Double Jeopardy Revisited
In any given time period, a small brand typically has far fewer buyers than a larger brand. In addition, its buyers tend to buy it less often. This pattern is an instance of a widespread phenomenon called
"double jeopardy" (DJ). The authors describe the wide range of empirical evidence for DJ, the theories that account for its occurrence, known exceptions and deviations, and practical implications.
I
N this article, we review a little-known but widely occurring and theoretically supported regularity in competitive markets, namely that small brands generally attract less "loyalty" among their buyers than large brands do among theirs. Twenty-five …show more content…
In allocating marketing effort it is worth knowing that it is normal for a small brand to attract somewhat less
"loyalty" and yet to survive. We return to these issues in the final discussion.
All the general explanations of double jeopardy are statistical, relating to the size structure of the market.
Other things being equal, small brands attract less loyalty just because they are small (i.e., have lower market shares). No other marketing mix or consumer variables need be invoked to explain DJ. McPhee himself outlined this notion in broad terms and it is also common to the later, more detailed models reviewed here.
What, then, is the role of marketing factors such
Journal of Marketing
Vol. 54 (July 1990), 82-91
as product formulation, price, distribution, advertising, promotions, and market segmentation? The answer seems to be that these factors give brands their different sales levels, which in turn show up in the
DJ pattern, but rarely cause big additional differences in brand loyalty. Hence competitive brands tend to differ mainly in how many buyers they have …show more content…
These brands obtain a somewhat higher average purchase frequency (about 20% or more) than their market shares or penetrations would warrant, in comparison with the w(l - b) or Dirichlet models (Ellis
1989). The explanation is not that repeat-buying loyalty for private labels is higher than normal, but that consumers ' access to any particular supermarket chain 's private label is necessarily restricted. In other words, it seems that the penetration b for private labels in the w(l - b) model is low rather than that the repeatbuying measure w is high; within their own stores, private label brands tend to have very high market shares. An occasional deviation seems to occur also for major market leaders. In our consulting experience, some leaders have a somewhat higher-than-expected level of repeat buying (i.e., they behave like an even bigger brand than they already are), but this deviation does not occur universally.
Summary Conclusions
Overall, the message is that frequent-purchase markets almost invariably show a regular or "lawlike" double jeopardy pattern whereby smaller brands tend to attract somewhat less loyalty, with deviations that are mostly relatively small and so far little understood. Not many such empirically and