What Marketing Research Can Actually Explain

Marketing AnalyticsEisend, M. · 2015Journal of Marketing
Topicsmeta-analysis·effect size·marketing knowledge·second-order meta-analysis·knowledge accumulation·pricing·consumer behavior

Your team cites "the research" to defend a major decision, and you must decide how much weight it deserves. Findings about pricing and consumer behavior explain more than findings about research methods or new product development. Which findings can you act on as they stand, and which need a local test first?

Marketing research findings are about as strong as psychology's.

Data chart

Marketing research findings are about as strong as psychology's

Average strength of the link between what studies tested and the outcome measured, pooled within each research field, from 0 (no link) to 1 (perfect link).

.24Marketing research.24Psychological, educational andbehavioral treatments.21Social psychology

Marketing findings are about as strong as findings in earlier reviews of psychological, educational and behavioral treatments and social psychology.

Marketing's .24 is rounded from .236 across 176 research reviews with more than 7,500 studies (1918-2012). The comparison values, .24 for treatments and .21 for social psychology, come from earlier research reviews cited in the Discussion (p. 36).

Data chart

Some areas of marketing research explain more than others

Share of the variation in the measured outcome that each area's research accounts for, in percent.

All marketing average: 6%Pricing13%Consumer behavior8%Strategy6%Channels6%Sales5%Advertising5%New product development4%Research methods2%
  • Pricing research explains the most outcome variation.
  • Judge research findings by subject area.

Findings about pricing and consumer behavior explain more outcome variation than findings about research methods or new product development, so judge research by subject area.

Table 5 (p. 32), averages within eight marketing research areas; the dashed line is the all-marketing average. Exact percentages: pricing 13.032%, consumer behavior 8.069%, strategy 6.128%, channels 5.657%, sales 5.219%, advertising 4.640%, new product development 4.111%, research methods 2.372%; overall 5.560%.

Action guide

  1. Use research findings to make marketing's case inside the firm.Quantify what research in the relevant area explains, to support the case for marketing's productivity, accountability and influence.
  2. Weigh research findings by subject area before a major decision.Ask your team how much research in that area explains about an outcome your decision depends on.
  3. Test findings about research methods and new product development locally.Before a major decision, ask your insights team whether these findings apply to your business.

Evidence

  • Marketing findings average a strength of about .24 on a scale from 0 (no link) to 1 (perfect link). Psychological, educational and behavioral treatments average .24, and social psychology averages .21.
  • Marketing's average finding explains less than 6% of outcome variation and leaves more than 94% unexplained. But an illustrative display with a 50% base rate puts success at roughly 38% without marketing knowledge and 62% with it.
  • Pricing research explains the most outcome variation, about 13%, and consumer behavior is second at about 8%. New product development explains about 4%, and research methods the least, about 2%.

Key takeaway

Marketing leaders can use relevant research findings to make marketing's case inside the firm.

Limitations

The studies span 1918-2012 across eight areas and measure how strong findings are, not financial returns or measured success rates in any market. Large effects may reflect simple questions, and immediate measures may miss lasting new product effects and advertising's carryover, the effect that lasts after the spend stops.

Source

Eisend, M. (2015). Have we progressed marketing knowledge? A meta-meta-analysis of effect sizes in marketing research. Journal of Marketing, 79(3), 23–40. https://doi.org/10.1509/jm.14.0288

https://doi.org/10.1509/jm.14.0288

Exact values behind the rounded figures: the pooled marketing correlation is .236; explained outcome variance is 13.032% for pricing, 8.069% for consumer behavior, 6.128% for strategy, 5.657% for channels, 5.219% for sales, 4.640% for advertising, 4.111% for new product development, 2.372% for research methods and 5.560% overall.

Read the paper ↗