The purpose of this experiment was to investigate the main and interaction effects of personalization (handwritten addresses versus computer-printed address label) and envelope color (brown versus white) on the response rate, response speed and response quality of a mail survey distributed to a business population. By examining multiple criteria of response speed and response quality as well as response rate, this study offers a more complete measure of effectiveness and expands on the majority of past research that has focused almost exclusively on the number of replies received. Moreover, it helps to redress the current paucity of research into the exploration of potential interaction effects among manipulated survey design features. No statistically significant main or interaction effects of personalization and envelope color on response rate, response speed or response quality were found. Practical implications of the findings for survey researchers are discussed. (c) 2006 Elsevier Inc. All rights reserved.