That's what I thought. So if our data says 3-10 days, we should plan (primarily) for the 10-end rather than the 3-end? Which was my suggestion -- that we cater to the 7-10-day visitors (and perhaps try to get them coming weekly if anything? (scheduled content, etc). > isaac wrote: > >I'm not sure what you're suggesting. > > sorry. > > what i'm suggesting is that the people who took the survey aren't > representative of all of our visitors and that it's most likely slanted > towards the side of "visiting more often". > > IOW, it's probably safe to say that the more often one visits evolt, the > more likely they are to have taken the survey. the less often one visits > evolt, the less likely they are to have taken the survey. > > i'm just suggesting that we make sure we understand these results by > considering the sample who took the survey. just that we keep it in mind > that, to some extent, these results are inherently flawed. understanding > HOW they are flawed will help us interpret them correctly. > > i explain that any better?