After the rapid and widespread emergence of online access panels, we are currently witnessing a new trend towards online custom panels that are specifically built, used and managed for research purposes of one company or its brand(s). This study compares the online access panel ‘XL Online Panels’ with a dedicated and branded online research panel from Heinz, generating conclusions on the relative advantages and disadvantages related to using either of them. Specific attention is paid to panel member conditioning and quality.

1. Prof. Dr. Kristof De Wulf Dr. Mike Friedman Bert Borggreve Pavlov revisited Effects of panel method & conditioning on panel member quality & product ratings Prof. Dr. Kristof De Wulf Dr. Mike Friedman Bert Borggreve

4. Each communication you have with a customer is also a door you open for them within your company. If you do not treat them with respect… they may not only decline to complete your survey, the may also take their business elsewhere. Dorsey 2007 Respect

5. Confidentiality

6. Fast and flexible

13. Method Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Wave 1 Wave 2 Wave 3 Wave 4 Wave 5 Heinz total sample n = 697 n = 770 n = 811 n = 1.082 n = 1.210 Heinz fresh sample n = 697 n = 223 n = 221 n = 356 n = 262 Heinz repeat sample n = 697 (1x) n = 553 (2x) n = 384 (3x) n = 288 (4x) n = 213 (5x) XLOP fresh sample n = 301 n = 213 n = 816 n = 218 n = 253

17. Testing panel method effects Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Wave 1 Wave 2 Wave 3 Wave 4 Wave 5 Heinz total sample n = 697 n = 770 n = 811 n = 1.082 n = 1.210 Heinz fresh sample n = 697 n = 223 n = 221 n = 356 n = 262 Heinz repeat sample n = 697 (1x) n = 553 (2x) n = 384 (3x) n = 288 (4x) n = 213 (5x) XLOP fresh sample n = 301 n = 213 n = 816 n = 218 n = 253

18. Testing panel method effects – Product ratings Product Rating (Unpriced Buying Intention) Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Sig. difference (95%) Effect size R 2 = .01 R 2 = .01 R 2 = .04 R 2 = .03

19. Testing panel method effects – Panel member quality Overall satisfaction questionnaire Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Sig. difference (95%) Fraction that filled in open ended question (likes) R 2 = .01 R 2 = .01 R 2 = .04 R 2 = .03 R 2 = .01 R 2 = .01 R 2 = .02 R 2 = .06 R 2 = .03

20. Testing panel method effects – Panel member quality Response rates between groups

22. Testing conditioning effects Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Wave 1 Wave 2 Wave 3 Wave 4 Wave 5 Heinz total sample n = 697 n = 770 n = 811 n = 1.082 n = 1.210 Heinz fresh sample n = 697 n = 223 n = 221 n = 356 n = 262 Heinz repeat sample n = 697 (1x) n = 553 (2x) n = 384 (3x) n = 288 (4x) n = 213 (5x) XLOP fresh sample n = 301 n = 213 n = 816 n = 218 n = 253

23. Testing conditioning effects – Product ratings Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Sig. difference (95%) Product Rating (Unpriced Buying Intention) R 2 = .01 R 2 = .01 R 2 = .03 Effect size

24. Testing conditioning effects – Product ratings – A zoom on all key scores of wave 5 Product Rating Wave 5: Cereal Sig. difference (95%) R 2 = .01 R 2 = .02 R 2 = .01 R 2 = .02 Effect size

25. Testing conditioning effects – Panel member quality Overall satisfaction questionnaire Wave 1: Soup Wave 2: Fruit drink Wave 3: Cereal Wave 4: Soup Wave 5: Cereal Sig. difference (95%) Fraction that filled in open ended question (likes) R 2 = .01 R 2 = .01 R 2 = .01 R 2 = .06

27. Effects of panel method

29. Effects of conditioning

32. Future directions