Research & Testing in Marketing – Test Group

View Github Repository Open presentation in a new window

jtcond13

See all presentation from jtcond13

Research & Testing in Marketing – Test Group

0 0

Research---Testing-in-Marketing

A reveal.js presentation on Direct Marketing Research

On Github jtcond13 / Research---Testing-in-Marketing

Research & Testing in Marketing

Jack Condon July, 2014

Direct Marketing Research

The purpose of direct marketing research is to predict likely buyers, identify likely defectors, find the next best product and to analyze the audience for our marketing.

Why?

To create rules about how an audience behaves. The easiest way to do this is by gathering consumer data.

  • Age
  • Income
  • Children
  • Length of Residence
  • Wealth
  • Homeowner or Renter
  • Married or Single
  • Credit Rating

Predicting Likely Buyers

Create two groups (preferably demographically similar) and send them a piece of marketing Then, take one of the groups and run a multiple regression or CHAID to create a predictive model Then, score both groups based on the model and sort the groups into deciles If deciles in both groups have a similar response rate then your model is good. Use this model to score prospects going forward

Test Group

Decile Number Buyers Buying 1 2600 397 15.27% 2 2600 230 8.85% 3 2600 113 4.35% 4 2600 93 3.58% 5 2600 76 2.92% 6 2600 44 1.69% 7 2600 24 .92% 8 2600 11 .42% 9 2600 9 .35% 10 2600 3 .12%

Control Group

Decile Number Buyers Buying 1 2600 400 15.38% 2 2600 237 9.12% 3 2600 109 4.19% 4 2600 107 4.12% 5 2600 72 2.77% 6 2600 41 1.58% 7 2600 20 .77% 8 2600 10 .38% 9 2600 8 .31% 10 2600 4 .15%

Identifying Defectors

Likelihood to defect is modeled just like marketing response (CHAID or multiple regression)

Sorting customers by these two variables will help identify where to focus retention efforts.

Firmographics

  • Great B2B Tool
  • Generally more detailed than consumer data
  • We can use this information to find firms that aren't buying products that similar firms are buying
  • Very helpful in understanding shift to defined-contribution (e.g. Darden, Sears moving to private exchanges first)

Audience Analysis

  • Attitudinal: Planners, Jugglers & Followers
  • Demographic: Geography, family status, etc.

Final Thoughts

  • It's a good idea to keep a record book of marketing testing so that knowledge about the audience can be shared.
  • E-mail: most important thing in e-mail marketing is the subject line. Test several subject lines before starting your campaign.

0