Advertising and Promotion Effects

Summary

Empirical generalizations on advertising, consumer promotions (coupons), and trade promotions from brand-level market response studies. Key findings: short-run advertising elasticity ≈ 0.10; advertising effects approximately double in the long run; 90% of effect dissipates within 6–9 months; price elasticity > advertising elasticity; retail promotions strongly increase short-run sales; coupon elasticity is very low (0.07).

Advertising Elasticity

Core Advertising Generalizations

Generalization [LS] (Leone & Schultz 1980): The short-term elasticity of selective advertising on own brand sales of frequently purchased consumer goods is positive but low.

Meta-analytic support: average short-run elasticity = 0.10–0.22 (Assmus, Farley & Lehmann 1984: 0.22; Sethuraman & Tellis 1991: 0.10; Lodish et al. 1995: 0.13).

Recommended null hypothesis for future models: 0.10 (conservative; BehaviorScan experiments suggest this is realistic for established brands).

Long-Run Effect

The average long-term effect of successful advertising spending is approximately double its initial effect.

Support: the long-run multiplier = $1/(1-\lambda )$. With $\lambda =0.5$, LRM = 2×. Aggregation-adjusted $\lambda =0.7$, LRM = 3.3×.

Duration Interval

The 90% duration interval for advertising of mature, frequently purchased, low-priced packaged goods is brief, averaging between six and nine months. [C]

Derivation: 90% of cumulative effect dissipates when:

$$T\ge \frac{\log (1-0.9)}{\log (\lambda )}=\frac{-1}{\log (0.7)}=6.5\text{ months}\text{\hspace{1em}(Eq 8.13)}$$

This is a short duration — contradicting common managerial beliefs that advertising effects persist for years. The finding implies advertising must be maintained continuously to sustain sales levels.

Factors Moderating Advertising Elasticity

Life Cycle Moderation

Advertising elasticities decline as a brand or market matures. [MBI]

Evidence (Shankar, Carpenter & Krishnamurthi 1999):

• Pioneer ethical drug brands: elasticity = 0.625
• Growth-stage entrants: 0.496
• Mature-stage entrants: 0.274

Lodish et al. (1995): new products: 0.26; established products: 0.05.

Implication: early-stage brands should invest heavily; diminishing returns set in as brand becomes established.

Other moderators identified by Millward Brown International:

1. Category development stage
2. Brand development level
3. Competitive context
4. Strategy effectiveness
5. Execution effectiveness

Advertising and Price Sensitivity

Three generalizations on the advertising-price interaction [KW]:

1. An increase in price advertising leads to higher price sensitivity among consumers.
2. An increase in feature advertising leads to higher price sensitivity among consumers.
3. An increase in non-price advertising leads to lower price sensitivity among consumers.

Managerial implication: price advertising and non-price advertising have opposing effects on price elasticity. Coordinating advertising and pricing decisions requires understanding which type of advertising dominates. Heavy non-price advertising can create price inelasticity and insulate a brand from price competition.

Consumer Promotions

Coupon Elasticity

Elasticity of coupon spending on own brand sales is positive but very low. [BG]

Average elasticity ≈ 0.07 (Bucklin & Gupta 1999, Hudson River Group ACNielsen analyses for Quaker Oats).

Corollary: Coupon response < media response. Coupons mainly attract the brand’s own heaviest users — who would have purchased anyway.

Implication: major packaged goods firms (P&G, Nestlé USA) have reduced coupon spending and shifted to targeted couponing at new product launches.

Retailer Sales Promotions

Temporary Price Reductions (TPR)

Temporary price reductions (TPRs) substantially increase sales. [BBF]

Deal discount elasticities greatly exceed regular price elasticities (Table 8-8):

• Tuna fish average: regular price elasticity −3.08 to −4.99; deal elasticity +8.1 to +10.3
• Margarine: regular −2.34 to −3.88; deal +4.84 to +8.65

Promotional price elasticities exceed non-promotional price elasticities. [BBF]

Trade Promotion Pass-Through

Retailers pass-through is generally less than 100% of trade deals. [BBF]

Retailers pocket some of the manufacturer-funded trade deal rather than passing it to consumers as price reductions. This partially explains why incremental sales from trade deals are often lower than expected.

Display and Feature Multipliers

Display and feature advertising have strong effects on sale items. [BBF]

Mean display multiplier for liquid dishwasher detergent: 1.69 (Crystal White Octagon: 1.5; Dove: 2.6; Ivory: 1.4; Palmolive: 1.5; Sunlight: 1.8).

Promotion elasticity categories (Narasimhan, Neslin & Sen 1996): higher in paper products, canned foods, pasta; lower in health & beauty aids and miscellaneous products.

BehaviorScan Advertising Weight Tests (Frito-Lay)

A large-scale set of 23 split-panel experiments over four years found:

• ~60% of TV commercials demonstrated sizable volume increases among advertised households
• Average gain: 15% among advertised households vs. matched no-advertising control
• Larger brands had lower probability of sizable gain (27% vs. 88% for smaller brands)
• “News” commercials (new products/line extensions) showed significant gains 100% of the time

Cross-Links

• Core generalizations framework: Marketing Generalizations Overview
• Carryover parameter ($\lambda$): Carryover Effects and Distributed Lags
• Price findings: Price and Distribution Effects
• Optimal advertising budget: Optimal Marketing Decisions and Forecasting
• Activity bias in advertising measurement: Activity Bias in Advertising
• Why RCTs (BehaviorScan, holdout experiments) are essential for measuring elasticities: Observational vs Experimental Methods in Advertising
• Experimental design principles behind BehaviorScan weight tests: The Experimental Ideal