How Much Personal Selling Moves Sales

ConversionAlbers, S.; Mantrala, M.K.; Sridhar, S. · 2010Journal of Marketing Research
Topicspersonal selling·sales force effectiveness·product life cycle·lingering sales effects·pharmaceutical detailing·sales benchmark

Imagine you are setting the budget for personal selling input, such as sales-force spending or sales calls, and deciding where to place it. Published studies disagree on how much sales respond, so a wrong benchmark can distort the budget. Which benchmark should your plan use, and which products and markets should get more selling input?

Personal selling has an adjusted benchmark for same-period sales.

Data chart

Where personal selling moves sales more

Predicted percent increase in same-period sales for a 1% increase in personal selling input (such as sales-force spending or sales calls), before adjusting for study design.

Product stageMature productsEarly-stage products0.271%0.535%Selling settingPharmaceuticalBusiness-to-business0.274%0.502%RegionUnited StatesEurope0.318%0.428%
  • Early-stage products respond about twice as much as mature ones.

A 1% increase in personal selling input moves same-period sales more for early-stage products than for mature ones, more in other business-to-business selling than in pharmaceutical detailing, and more in Europe than in the US.

Predictions from 506 estimates in 75 studies (56% pharmaceutical, 26% defense recruiting) in the US and Western Europe, over about four decades up to 2010, each holding every other study feature at its average; exact values .535 and .271, .502 and .274, .428 and .318. Other business-to-business selling covers supplier sales to retailers, newspaper advertising space sales and industrial goods sales. The regional gap rests on the weakest evidence of the three.

Data chart

Sales respond less to personal selling in more recent data

Predicted percent increase in same-period sales for a 1% increase in personal selling input, by the year the study data were collected, before adjusting for study design.

Data from 19820.351%Data from 19920.304%Data from 20020.26%

The more recent the data, the smaller the predicted same-period sales response to a 1% increase in personal selling input.

Predictions at three data-collection years from the same 506 estimates, each holding every other study feature at its average; the predicted response falls by about .005 per year of data; exact values .351, .304 and .26.

Action guide

  1. Use personal selling's adjusted average sales response as your starting benchmark.Before budgeting, have your analytics team check each estimate for market fit and design; older data tend to give higher estimates.
  2. Give early-stage products more of your sales force than mature ones.Their predicted sales response is about twice as high.
  3. Favor Europe over the US for selling when margins and current sales are comparable.Predicted sales response to selling is higher there.
  4. For mature pharmaceutical products, test complementary channels before adding detailing.Sales respond less to detailing than to other business-to-business selling.
  5. Adjust published estimates to one response model before comparing them.Have your analytics team do this before estimates shape your sales-force budget.
  6. Require separate same-period and later-period sales models.Have your analytics team state how much of the response it assumes carries over, since same-period response is not the whole.

Evidence

  • On average, a 1% increase in personal selling input goes with about 0.31% higher same-period sales. That figure adjusts each study's estimate for design choices that push it up or down, such as leaving out promotions or past selling. At that level, personal selling remains a relatively strong part of the marketing mix in many industries.
  • A 1% input increase predicts about 0.54% more same-period sales for early-stage products versus 0.27% for mature ones, before adjusting for study design.
  • A 1% input increase predicts about 0.43% more same-period sales in Europe versus 0.32% in the US, before adjusting for study design.
  • A 1% input increase predicts about 0.27% more same-period sales for pharmaceutical detailing versus 0.50% for other business-to-business selling, before adjusting for study design.
  • Sales respond less to selling in more recent data. Before adjusting for study design, a 1% input increase predicts about 0.35% more same-period sales in data from 1982, 0.30% from 1992 and 0.26% from 2002.
  • In models that track later-period effects, an average 0.76 share of the sales response carries into the next period. It is not a long-run sales response.

Key takeaway

Use personal selling's adjusted same-period benchmark, then adapt it for product stage, market and study design.

Limitations

The evidence mainly covers pharmaceutical and defense-recruiting personal selling in the US and Western Europe over about four decades up to 2010. It measures sales response, not profit, which also depends on margins and the cost of selling.

Source

Albers, S., Mantrala, M. K., & Sridhar, S. (2010). Personal selling elasticities: A meta-analysis. Journal of Marketing Research, 47(5), 840–853. https://doi.org/10.1509/jmkr.47.5.840

Exact values behind the rounded figures: a sales response of approximately .31 after adjusting for study design (.34 before); predicted responses of .535 for early-stage and .271 for mature products, .428 for Europe and .318 for the US, .274 for pharmaceutical detailing and .502 for other business-to-business selling, and .351, .304 and .26 in data from 1982, 1992 and 2002.

Read the paper ↗