Back in the day “hits” to one’s website was the only metric we had. We quickly learned that hits were pretty meaningless, given that a single visitor looking at a single page might generate several hits by doing so, because each image, block of text, script, CSS or other chunk of data is counted as a hit.

We then learned that visitors are what counts, but even that distorted one’s perception of a site’s popularity because a single visitor might visit more than once. The concept of unique visitors came to the fore, so that only a the viewing of one’s site as a whole by a visitor who had not been to the site within a given timeout period from the same computer counted.

Then along came social media, where visitors from Twitter, Facebook, MySpace, LinkedIn etc etc etc suddenly became the more important visitors. But, still we simply added those visitor numbers to the overall “score”.

But.

One thing that social media has taught us is that it’s the individual that counts (although they may be part of a crowd). And, if the individual is all that counts, then what they do on your website when they visit is all important. Do they visit ever so fleetingly and disappear, perhaps via a link to another site or via an AdSense ad? Do they grab your newsfeed or subscribe to your email newsletter? Follow you on Twitter or Like you on Facebook? Does any of that matter?

We all chase the numbers, the visitor rates, the crowds, but good analytical software will let you drill down to see whether your website is really engaging its visitors. One way to do this drilling is to filter out all visits that lasted less than 30 seconds. 30 seconds is probably the minimum length of time you could say a visitor was engaged if they hit your site. Any shorter and they’re just surfing old-style but have a slow connection.

I recently ran such a filter using Google Analytics and was not entirely shocked to discover that despite the well over a million visits to the site (Sciencebase) during 2010 it was only about 1 in 3 of those visits that lasted more than 30 seconds. This is actually quite reassuring and does explain my fairly reasonable AdSense revenues. By the way, it’s only visitors who land on the site from a search engine who see the main block of AdSense ads, if you’re engaged via another site, a bookmark, or the site’s newsfeed, you shouldn’t see those at all and can enjoy the site unhindered by flagrant commercialization.

Actually, 1 in 10 is a very good marketing uptake (not that I’m really selling anything other than free information on the site), so knowing that one out of every three visitors to Sciencebase is spending quality time with the site is even more gratifying. A similar picture is painted when I filter Sciencetext stats in this manner.

Indeed, on average all visitors during 2010 spend 50 seconds on Sciencebase but those “engaged” visitors average almost five minutes. The average visitor looks at 1.34 pages but the “engaged” visitor reads 3 pages or so before heading off to the next site.

This post was inspired by a recent article by Chris Rand on filtering Analytics, by Andrew Maynard’s Facebook revelations regarding his site’s traffic in 2010 and as a riposte to an article suggesting that one shouldn’t use AdSense at all because it’s wasting visitor engagement.