Any of you who use Blogger to host your blog probably use Google Analytics, Google's free online report center that gives you information on what people are reading on your site, how many are viewing your site and where the traffic is coming from. But Google Analytics is seriously flawed in a few of it's statistics, flaws that have bugged me off and on over the last two years, flaws that I finally discovered the reasons behind and now it only ends up annoying me even more! Sure Google Analytics is free, so complaining about a free service is like complaining about a free meal, but you still ought to be aware of these issues because if you take Analytics stats at face value you might be disappointed or incorrectly enthusiastic about what you see. Plus I just don't get how Google couldn't fix something that's so blatantly screwed up.
If you take a look at the screen capture above you'll probably see a few stats that you might see every once in a while on your own account, I'm speaking of the 00:00:00 on the Avg. Time on Page column. Now for a post that has double digit hits and a couple comments there should be no 00:00:00, in fact even if someone simply clicked on a link to a certain post and scrolled to the bottom that would take a couple seconds, therefore 00:00:00 should never be the average! But Analytics is screwed up, and this is what I discovered. Analytics doesn't record the time spent on a certain post until the visitor has clicked on another link within in your blog. So if a visitor spends 20 minutes reading your post on Top 10 Movies then exits your site Google Analytics logs that as 00:00:00, if that visitor clicks on link which leads them to another post within your site Analytics will then log the 20 minutes, but if they exit on that second post the second post gets a 00:00:00. See the absurd pattern?
Now what really aggravates me about this issue is if Analytics can log a visitor leaving your site why can't it also log the time on the page they exited on? If you refer to the screen shot above you'll notice my post with the most views had an exit rate of 100%, that means everyone who viewed that post exited my site after viewing it. If half those people went on to read another post the rate would be 50% (like the 3rd post from the top) So if Analytics knows when they leave in order to calculate the exit % why can't it time stamp the time when they left that page? Why does it do it only when someone clicks another link within the site? So essentially, Google Analytics average time on page per post is wrong, all the time, and therefore the average time on site is also wrong all the time. So if your confused is to why the time stamps seem pretty low don't be, while the biggest innovator on the internet should be able to fix this blatant issue it's obvious they don't care and are content providing people with extremely false information that could potentially discourage some people.
Now the other statistic in question doesn't really bother me, it's simply a pointless statistic to have on here, it's the page views column. The page views stats is exactly like the page counters people put on there websites to somehow show how popular their site is; they're worthless. Why? Because I could go on someone's blog pick a particular post and hit refresh 50 times and on their stats for that day that post would show 50 page views and 1 unique page view. The unique informs you how many different visitors viewed that post during that day, while the regular page views signifies how many times that post was loaded, so page views and most online counters are only extremely bloated numbers and do not correctly reflect the legitimate and accurate traffic on your site.
So what's the point of Google Analytics? Well aside from the fact you're better off ignoring anything related to time and regular page views, Analytics is a decent source for determining what posts are popular on your site (unique hits) and search engine hits that are bringing people to your site, that's really the only useful information Analytics brings you which can help you in determining what you might want to write about in the future. A few of the other stats, like where people are from are interesting but not too helpful. Overall, I really wish you could manually delete the average time on page column because it's incredibly misleading to those who don't understand how Google calculates and incredibly frustrating to me when I don't understand why Google can't or won't fix it because it is so misleading, and if it worked like it should it would be quite useful. What do you think?
Interesting. I usually just check google analytics once in a blue moon to see what I'm bringing in search wise. I stopped paying attention to the "Time on Site" thing because it's always seemed a big screwy to me. I doubt GA is high on their priority list, especially if there's nobody of high importance, from the companies perspective, complaining about it (tends to be how support goes).
ReplyDeleteThanks for doing the research though, it does explain a lot :).
Wow I've wondered about the "average time spend on page" stat since it always seems a little screwy. What a weird bug to have! Thanks for letting us know about it, though, it'll certainly change how I look at my stats!
ReplyDeleteI didn't get half of this quantum physics of yours :)))
ReplyDeleteMy site gets around 800 to 1200 visits each day, out of which around 70%-85% are unique ones.
@Univarn - well it's definitely not worth paying for, I think they have a paid version, or you have to start paying for it once you get up to a certain number of hits per year. I check it daily, mainly for the search engine hits and to see which reviews are getting the most attention.
ReplyDelete@Alex - no problem.
@Dezmond - it's not really that difficult to understand, basically any of the average time stats aren't even worth looking at because they're all wrong.