Television Advertising Weight and Sales Response

ConversionLodish, L.; Abraham, M.; Kalmenson, S.; Livelsberger, J.; Lubetkin, B.; Richardson, B.; Stevens, M. · 1995Journal of Marketing Research

Topicstv advertising·advertising weight·split cable·grp·advertising elasticity·copy testing·cpg

An annual TV plan assumes more exposure will grow an established packaged-goods brand. Advertising weight means how much advertising you run. Imagine you are approving next year's increase: should it require fresh sales evidence?

More TV advertising is not a dependable way to grow sales for established packaged-goods brands.

Data chart

New-product television increases lift sales more often

Percent of television advertising tests whose heavier plan gained sales volume or market share at the study's 80% confidence threshold. The benchmark is the share expected to pass by chance alone.

A clear gain is less common for established brands.

TV increases for new products showed a clear sales gain more often than those for established brands.

A test counts when the heavier plan's sales-volume or market-share gain reaches the study's 80% confidence threshold; 20% would pass by chance alone. Positive results were 33% of 207 established-brand tests and 55% of 85 new-product tests (p. 128).

Data chart

More television advertising did not improve the odds

Average percent increase in television exposures to the target audience in the heavier plan over the regular plan, for established-brand tests with and without a clear sales gain.

Bigger exposure increases did not distinguish successful tests.

For established brands, tests with a clear sales gain had smaller average exposure increases than tests without one.

Exact averages +77% and +88% from Table 2 (p. 128), which covers 141 tests with exposure data across new and established brands. The table gives no separate established-brand count for these rows; its column heading is not that subset's count. A clear gain uses the same sales-volume or market-share threshold as Visual 1.

Action guide

Require sales evidence before adding TV weight to an established brand.Your insights team must report incremental sales from an in-market test.
Set separate response assumptions for launches and established brands.Ask your media team to test more prime-time exposure and a greater share of advertising weight at launch.
Test TV increases alongside changed copy strategy and penetration objectives.Have your brand team compare sales before approving more weight.
Judge established brands' TV spending on in-market sales, not recall or persuasion scores.Those scores have no clean cutoff for success.
Pilot TV cuts in test markets before cutting nationally.This plan is untested. Cut nationally only if sales there hold over time. Keep comparing old and reduced weight, and restore the old level if sales fall. Test new copy meanwhile.

Evidence

Only 33% of TV weight tests for established brands showed a clear sales gain, against 55% for new products. Bigger increases did not raise the odds for established brands.
For established brands, a 1% increase in TV weight went with about 0.05% more sales volume, not clearly different from zero. The sample may overstate this response.
For established brands, changed copy strategy, attitude-changing messages and a penetration objective went with larger sales effects.
TV increases were more likely to lift sales in growing or frequently purchased categories and where in-store merchandising was light.
For established brands, advertising in bursts separated by gaps went with smaller sales gains. For new products, bigger gains went with more prime-time exposure and a greater share of weight at launch. Adding weight at the start or end of a plan was more likely to lift sales than spreading it throughout.
For established brands, standard recall and persuasion scores showed no strong relationship with sales response and no clean cutoff for success.

Key takeaway

Test added TV weight for established packaged-goods brands before relying on it to grow sales.

Limitations

Television tests of frequently bought, low-priced U.S. packaged goods in 1982-1988 measure sales volume and share, not profit or long-term brand equity. They do not establish how digital media, other sectors or other markets respond.

Source

Lodish, L., Abraham, M., Kalmenson, S., Livelsberger, J., Lubetkin, B., Richardson, B., & Stevens, M. (1995). How T.V. Advertising Works: A Meta-Analysis of 389 Real World Split Cable T.V. Advertising Experiments. Journal of Marketing Research, 32(2), 125-139. [Published paper](https://journals.sagepub.com/doi/10.1177/002224379503200201).