Amongst general panellists who are past 12-month active respondents, response rates can vary from 20% to 60% depending on the topic, incentive, population, subgroup, survey length, etc. Although response rates vary, a low response rate can still produce a representative sample. We employ a two-stage approach to ensure representativeness. In the first stage, we balance the outgoing sample demographically before survey invitations are delivered to panellists. In the second stage, we weight the resulting data we collect by a variety of factors, including but not limited to, age, sex, education, and income, region of the country, Internet usage and shopping behaviour. These two stages combined with our statistical process known as propensity score weighting helps to ensure that the resulting survey data are representative of the populations of interest to clients. In addition, our adherence to a balanced sampling methodology and best-in-class panel practices allow us to maintain low drop-out and suspend rates. Response rate is calculated as the number of responses (completes, suspend, terminate/non-qualified completion) divided by total number of invitations sent.