Abstract
Many markets have historically been dominated by a small number of best-selling products. The Pareto principle, also known as the 80/20 rule, describes this common pattern of sales concentration. Several papers have provided empirical evidence to explain the Pareto rule, although with limited data. This article provides a comprehensive empirical investigation on the extent to which the Pareto rule holds for mass-produced and distributed brands in the consumer-packaged goods (CPG) industry. We used a rich consumer panel dataset from A.C. Nielsen with 6 years of purchase histories from over 100,000 households. Our analysis utilizes a large number of potential factors such as brand attributes, category attributes, and consumer purchase behavior to explain variation in the Pareto ratio at the brand level across products. Our main conclusion is that the Pareto principle generally holds across a wide variety of CPG categories with the mean Pareto ratio at the brand level across product categories of .73. Several variables related to consumer purchase behavior (e.g., purchase frequency and purchase expenditure) are found to be positively correlated with the Pareto ratio. In addition, niche brands are more likely to have a higher Pareto ratio. Finally, brand/category size, promotion variables, change-of-pace brands, and market competition variables are negatively correlated with the Pareto ratio.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
For a robustness check, we conducted the same analyses on the basis of the original definition created by A.C. Nielsen and the results were not significantly changed (please refer to the footnote 6 for details).
As a second robustness check, we conducted analyses using a volume definition of the Pareto ratio in addition to the dollar sales definition. Using a volume definition, the mean of the Pareto ratio is .72 compared to the Pareto ratio measured by dollars (.73).
The list of product categories used in the analyses is the following: cigarettes, carbonated soft drinks, low-calorie soft drinks, toilet tissue, nutritional supplements (vitamin), cookies, ice cream (bulk), canned soup, candy-chocolate, wine, ground and whole bean coffee, yogurt, bottled water, liquid detergent, frozen pizza, potato chips, fruit drinks (canned), light beer, paper towels, orange juice, cheese, and cereal (ready to eat).
Households with at least five purchases in a category were labeled as “category users” and were included in the analyses. We also did some sensitivity analysis on the threshold to define a category of users by altering a cutoff point (i.e., 1, 3, 5, and 10). The Pareto ratio slightly increases under the cutoff point at 1 and 3, but the results from 5 and 10 are the same.
Additionally, we did not include the brands with less than 1% market share because including those brands in our analysis would inflate the PR, which would be higher than the current number. In addition, in the CPG market, there are many small brands that did not exist during our whole data collection period (i.e., 2004–2010), so focusing on main brands (which have existed in the market over the long term) would show clearer patterns of the PR.
The mean of PR at the brand level based on the original definition of A.C. Nielsen is .65, which is not significantly different.
In this paper, the top 20% of consumers was determined by an amount of total dollar expenditure for a certain brand during the observation period. We defined the top 20% of consumers by sorting all consumers of each brand on the basis of total dollar expenditure for each brand. The top 20% of consumers could be either frequent shoppers or large basket shoppers since total dollar expenditure is a function of basket size and purchase frequency. For example, total dollar expenditure could be higher if consumers frequently purchase the brand although expenditure per shopping trip might be small. On the other hand, this number could also be higher if the basket size is large, meaning that expenditure per shopping trip is large although consumers rarely purchase a particular brand. Thus, it could be controversial to define the top 20% of consumers by their behavioral loyalty. However, interestingly, in our dataset, consumers having a higher purchase frequency were also more likely to have a larger basket size. We concluded that the top 20% of consumers, based on the total dollar expenditure, were behaviorally loyal consumers and we will investigate correlates (i.e., that have been studied in previous literature to see the effects of those on brand loyalty) to see how those affect the PR.
For the robustness check, we conducted the same analysis with larger samples including 16,000 brands from 100 product categories. The results were consistent with the results from smaller samples.
We also ran a regression with different samples, separated by an observation period, to see the difference between samples. Since there might be a difference between consumers who had been on the panel for 6 years and for 1 year, we checked the robustness of our regression results by separating observations with the sample length. First, we ran a regression with observations only from 2004 to 2006 and from 2007 to 2009, respectively. Next, we ran a regression of the total observations from 2004 to 2009, and then compared the results between those three regressions. The results are consistent with the prior findings.
We also looked at the correlations between the independent variables. For both brand- and category-level attributes, most variables are not correlated with each other except the purchase (and promotion) frequency and expenditure. As we mentioned in footnote 7, in our data, the purchase (and promotion) frequency and expenditure are highly correlated, so to deal with multicollinearity, we ran multiple regressions with and without those correlated variables.
We also repeated the same analyses with continuous independent variables. In order to capture any non-linear relationships, we included quadratic terms of the continuous independent variables such as market share, brand penetration, purchase frequency/expenditure, and promotion frequency/expenditure. The results are consistent with the discretized ones (except the coefficient of the niche-brand dummy became insignificant).
References
Aaker, D. A. (1991). Managing brand equity. New York: The Free Press.
Anderson, C. (2006). The long tail: why the future of business is selling less of more. New York: Hachette Books.
Brynjolfsson, E., & Smith, M. (2000). Frictionless commerce? A comparison of internet and conventional retailers. Management Science, 46(4), 563–585.
Brynjolfsson, E., Hu, Y. J., & Smith, M. D. (2003). Consumer surplus in the digital economy: estimating the value of increased product variety at online booksellers. Management Science, 49(11), 1580–1596.
Brynjolfsson, E., Hu, Y. J., & Simester, D. (2011). Goodbye Pareto principle, hello long tail: the effect of search costs on the concentration of product sales. Management Science, 57(8), 1373–1386.
Dodson, J. A., Tybout, A. M., & Sternthal, B. (1978). Impact of deals and deal retraction on brand switching. Journal of Marketing Research, 15(1), 72–81.
Ehrenberg, A. S. C (1972). Repeat-buying : theory and applications. Amsterdam: American Elsevier.
Ehrenberg, A. S. C. (1988). Repeat-buying: facts, theory and applications, 2nd ed. Edward Arnold, London; Oxford University Press, New York. Reprinted in the Journal of Empirical Generalisations in Mark Science, 2000, 5, 392–770 (www.empgens.com).
Ehrenberg, A. (2000). Repeat buying. Journal of Empirical Generalisations in Marketing Science, 5(2).
Ehrenberg, A. S. C., Goodhardt, G. J., & Patrick Barwise, T. (1990). Double jeopardy revisited. The Journal of Marketing, 54(3), 82–91.
Ferrari, S., & Cribari-Neto, F. (2004). Beta regression for modelling rates and proportions. Journal of Applied Statistics, 31(7), 799–815.
Gupta, S. (1988). Impact of sales promotions on when, what, and how much to buy. Journal of Marketing Research, 25(4), 342–355.
Kahn, B. E., Kalwani, M. U., & Morrison, D. G. (1988). Niching versus change-of-pace brands: using purchase frequencies and penetration rates to infer brand positionings. Journal of Marketing Research, 25(4), 384–390.
Krishnamurthi, L., & Raj, S. P. (1991). An empirical analysis of the relationship between brand loyalty and consumer price elasticity. Marketing Science, 10(2), 172–183.
McPhee, W. N. (1963). Formal theories of mass behaviour. New York: The Free Press of Glencoe.
Morrison, D. G., & Schmittlein, D. C. (1981). Predicting future random events based on past performance. Management Science, 27(9), 1006–1023.
Morrison, D. G., & Schmittlein, D. C. (1988). Generalizing the NBD model for customer purchases: what are the implications and is it worth the effort? Journal of Business and Economic Statistics, 6(2), 145–159.
Neslin, S., & Shoemaker, R. W. (1989). An alternative explanation for lower repeat rates after promotion purchases. Journal of Marketing Research, 26(2), 205–213.
Raj, S. P. (1985). Striking a balance between brand “popularity” and brand loyalty. Journal of Marketing, 49(1), 53–59.
Rosenberg, L., & Czepiel, J. (1983). A marketing approach for consumer retention. Journal of Consumer Marketing, 1(1), 45–51.
Schmittlein, D. C., Cooper, L. G., & Morrison, D. G. (1993). Truth in concentration in the land of (80/20) laws. Marketing Science, 12(2), 167–183.
Shin, J., & Sudhir, K. (2010). A customer management dilemma: when is it profitable to reward one’s own customers? Marketing Science, 29(4), 671–689.
Twedt, D. W. (1964). How important to marketing strategy is the heavy users? Journal of Marketing, 28(1), 71.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kim, B.J., Singh, V. & Winer, R.S. The Pareto rule for frequently purchased packaged goods: an empirical generalization. Mark Lett 28, 491–507 (2017). https://doi.org/10.1007/s11002-017-9442-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11002-017-9442-5