> The author here takes the average of the conversion rates from different websites. Am I wrong in saying that isn't quite right? Yes. It gives too much weight

> The author here takes the average of the conversion rates from different websites. Am I wrong in saying that isn't quite right? Yes. It gives too much weight to sites that referred a smaller number of people, and too little to sites that referred a larger number. > Shouldn't the "average conversion rate" be just the sum total of users that go to Toggl and sign up over the total number of users that went to Toggl? Not "users that go to Toggl", but "users that are referred to Toggl from one of those sources", on both sides of the fraction. (Possibly you knew this, but it seemed worth clarifying.) Equivalently, a weighted average of the conversion rates, weighted by the number of referrals from each source. > it just seems like an opportunity to inflate your conversion rate numbers. Note that this could deflate numbers just as well as inflate them.