This looks fishy. These guys obviously have a conflict of interest with regard to the study, as they're trying to sell you their marketing analysis services.

This looks fishy. These guys obviously have a conflict of interest with regard to the study, as they're trying to sell you their marketing analysis services. This reminds me of the studies conducted on google adwords, where research companies said that 30% of clicks were clickfraud. They arrived at this number by getting a group of ad purchasers together and asking them what the number was. They guessed, and their guesses were averaged. I think another study said the number was closer to 80%, but again their methodology was poor. I don't buy this. That this study covers two million people doesn't mean it is representative, because these were people that filled in an online survey and talked about their attitudes and intentions. They missed a huge part of the internet users who, by and large, have better things to do, and don't fill out surveys.