Long-Term Effects of Price Promotions

ConversionPauwels, K.; Hanssens, D.; Siddarth, S. · 2002Journal of Marketing Research
Topicsprice promotions·long-term effects·persistence modeling·var·category incidence·brand choice·purchase quantity·scanner data

Imagine your national soup brand's team wants to renew retailer promotions, arguing they build lasting market share. A promotion adds sales when more shoppers buy soup (category buying), more of them pick your brand (brand choice) or each buys more. Should that promise decide the renewal, or the weeks around each promotion?

Price promotions are associated with short-lived sales gains and virtually no permanent gain. For national brands, category buying leads the total response.

Data chart

Permanent promotion effects are rare, and most of the response arrives within weeks

Percent of brand-store cases with permanent effects on brand choice, then average and maximum weeks until 90 percent of the measurable promotion response has arrived.

3%Brand choice cases with permanentpromotion effects2 weeksAverage time until most of themeasurable response arrives8 weeksMaximum time until most of themeasurable response arrives
  • Category buying and purchase quantity show no permanent promotion effects.

Permanent effects are absent for category buying and purchase quantity and rare for brand choice.

Permanent effects occur in 1 of 29 brand-choice cases, reported as 3% (p. 431). Timing covers 26 analyzed brand-store combinations and measures arrival of 90% of the measurable response, not its complete disappearance; zero weeks means that share arrives in the promotion week. Exact component averages in Table 7 (p. 433): soup category buying 1.92 weeks, brand choice 2.75, purchase quantity 2.38; yogurt 2.53, 1.70 and 1.47 weeks, respectively.

Data chart

Category buying leads once the following weeks are counted

Percent of a price promotion's measured sales response from more shoppers buying the category, shoppers choosing the promoted brand, and larger purchases, in the promotion week and with following weeks added.

Soup promotion weekCategory buyingBrand choiceLarger purchases30%48%22%Soup with following weeksCategory buyingBrand choiceLarger purchases66%11%23%Yogurt promotion weekCategory buyingBrand choiceLarger purchases14%73%13%Yogurt with following weeksCategory buyingBrand choiceLarger purchases58%39%3%
  • Category buying leads when the following weeks are counted.

Category buying overtakes brand choice in the total response for both categories, though not for private labels.

Printed shares from Table 12 (p. 436), averaged across the analyzed stores: soup promotion week 30% category buying, 48% brand choice, 22% larger purchases; promotion plus following weeks 66%, 11%, 23%; yogurt promotion week 14%, 73%, 13%; promotion plus following weeks 58%, 39%, 3%. These are category averages; private labels do not follow the category-buying lead. The printed total estimates do not all reconcile with the immediate and adjustment estimates in Tables 6 and 8. Soup's total brand-choice share is 28% and 20% in the alternative model and weighting checks (footnote 11, p. 436), compared with 11% here.

Action guide

  1. Keep promoting while temporary sales gains are profitable.Judge sales and profits during and up to two months after each promotion, and plan for no permanent growth.
  2. Measure category buying and brand choice separately.Brand choice leads in the promotion week, but category buying leads for national brands once the following weeks are counted.
  3. Separate the retailer's category gains from your brand's profit.Do this when you renew promotion deals you fund. Counting the following weeks, category buying leads for national brands, brand choice for private labels.

Evidence

  • Category buying and purchase quantity show no permanent promotion effect, and only 1 of 29 brand-choice cases shows one, which implies no structural damage to sales. Tracking sales and profits during and up to two months after each promotion is enough, and promoting looks better than stopping while that window is profitable.
  • The time until 90% of the measurable sales response has arrived ranges from 0 to 8 weeks. It averages about two weeks across soup and yogurt brands.
  • In the promotion week, brand choice is the biggest source of extra sales, 48% for soup and 73% for yogurt on average.
  • After the promotion week, extra category buying continues while part of the brand-choice gain reverses, so category buying averages 66% of soup's total response and 58% of yogurt's. Profitable promotions may benefit retailers more than manufacturers, which could explain why retailers press manufacturers to promote.
  • A 1% unexpected price cut is associated with about 5.93% more category buyers for national soup brands and about 1.01% fewer for private labels, counting the following weeks.
  • A 1% unexpected price cut is associated with about 5.45% more category buyers for national yogurt brands and about 0.71% more for private labels, counting the following weeks.

Key takeaway

Price promotions are associated with short-lived sales gains and virtually no permanent gain.

Limitations

The study covers store purchases of canned soup and yogurt in a mature US market during 1986-1988 and measures sales, not profit or margins. It does not test growth categories, other channels or other countries.

Source

Pauwels, K., Hanssens, D. M., & Siddarth, S. (2002). The Long-Term Effects of Price Promotions on Category Incidence, Brand Choice, and Purchase Quantity. Journal of Marketing Research, 39(4), 421-439. https://doi.org/10.1509/jmkr.39.4.421.19114.

Exact values behind the rounded figures: average response times for soup are 1.92 weeks for category buying, 2.75 for brand choice and 2.38 for purchase quantity; yogurt averages are 2.53, 1.70 and 1.47 weeks, respectively.