Since 2010 GESIS has been conducting a pilot study to develop a strategy for recruiting and maintaining an online access panel based on a probability sample. The target population were German-speaking adults living in private households in Germany who identified themselves as using the Internet for non-work related purposes. Panel members were recruited by means of a telephone interview. Samples were drawn implementing a dual frame approach (i.e., including both landline and mobile numbers).  Internet-users were requested to join an online panel for which they had to provide their email address. In three studies nearly 1800 respondents were willing to join the panel. These respondents were subsequently invited to 8 online surveys, one survey per month. In the panel we replicated many questions from ALLBUS, ESS, Eurobaromater, SOEP, including a large set of demographic variables. The pilot implemented various experiments to optimize recruitment strategies. We present response rates and consent rates with regard to incentives, interview opening, and performance differences of interviewers and fieldwork agencies. Furthermore, the longitudinal design allowed us to look at changes in response behaviour due to continued participation. We present findings on aspects of panel conditioning and respondent burden. In summary, we find substantive and high effects across different recruitment strategies but rather low or few negative effects of continued participation in the panel.