Most of the audience numbers you read about for ad-supported sites are vastly overstated. Here’s why.
Let’s say a site says it has 1 million monthly users. That means, of course, that a million people are estimated as having gone to the site at least once in the month (or the numbers may come off their server records).
However, not all visitors are created equal. In fact, generally on the best sites, only 20-30% of the users do 80-90% of the viewing, whether measured by time spent or by number of pages viewed per visit. So only 2-300,000 viewers of the million are “real”.
The other 70-80% do very little of the viewing and a lot of them visit only once, probably through a search engine, and didn’t even really know what site they were on. These visitors are almost worthless on an ad-supported site. Why? Because they only see an ad once, an exposure that makes little or no impression.
It’s always been shown in any medium that an ad needs to be seen multiple times before it sinks in and causes action. This theoretical site can really only give advertisers a good return for their money against the frequent visitors.
Some sites have even less core users, with 90% or more in the single page category. There was a major search engine that is no longer a factor that had an amazingly small number of repeat visitors. Almost all of them were one-timers. That profile predicted the demise of the site and it did indeed die.
Another frequent trick is to report “registered users”. These are people who have signed up with a log-on and password. These numbers can be really misleading, since many people sign up for an account and then never use it. Or they have multiple accounts. What you need to know is how many registered users accessed their account at least once in the month.
How many users does Facebook or Twitter really have? Hard to say, but my dog is one of them.