Response rates to business surveys are often quite low. Because of the special circumstances and characteristics of business surveys, average responses rates to business surveys may range from lows of 10% to no higher than about 50%. We were interested in understanding whether business respondents had preferences for different survey modes. In a recent mail survey of businesses in Washington State, we included a question, asking business respondents what their most preferred survey mode was. The response options included the following: mailed survey, telephone survey, web survey, face-to-face survey, and no preference. The sample of almost 3,000 businesses was stratified by geographic region, business size, and industry. The mail survey was implemented using total design methods (TDM) to encourage a high response rate. The survey included an internet response option, and also a telephone followup of nonrespondents. The telephone followup attempted to get information from businesses on their survey mode preferences. A total of 1400 businesses responded to the survey, with 1150 responding by mail, and 250 responding on the web. Our paper presents an analysis of the survey mode preferences by business strata. We also compare the results of survey mode preferences for respondents and nonrespondents. The mode preferences of business responding by mail and those responding by web are different, and our paper presents thesedifferences. We discuss the implications of these survey mode preferences for improving response rates to business surveys in the future.