Each year the Consumer Reports National Research Center conducts a large-scale survey, which asks Consumer Reports magazine and online subscribers to report on their experiences with a variety of products (e.g. automobiles, electronics) and services (e.g. HMOs, cell-phone service). Both print and online survey methodologies are used. Generating close to one million responses a year, our Annual Questionnaire is one of the largest surveys in the world. Additionally, CRNRC conducts several other online surveys throughout the year. Maintaining high response rates to these surveys is vital to the organization as surveys are an important source for magazine and Web-site content. In turn, over the past several years, we have conducted a variety of tests examining the impact of various factors (e.g., invitation wording, questionnaire appearance, domain names, incentives) on response rates. Due to the large number of subscribers we survey each year, we have been able to test the effects of multiple different manipulations. Our paper will present which tests have been most and least effective and we will discuss similarities and differences between print versus online modes. For example, we will share data which suggests that incentives that are effective in increasing print survey response rates are largely ineffective when it comes to online surveys. We will also focus on various issues unique to online surveys (e.g., the impact of small changes in email subject headers, the importance of domain names coming from “trustworthy” sources).The discussion will focus on how our findings are similar to and different from those already discussed in the literature (and what might explain differences). In addition we will speak to the larger conference theme in addressing the unique challenges faced by groups such as Consumer Reports for whom surveying provides a public service but for whom dropping response rates are an increasing concern.