Is Google Analytics Accurate for Counting Unique Hits?
We all know that Webmasters are generally a lazy bunch. The same holds true for coders, another lazy bunch. And I mean that in an affectionate way because I too am guilty of being lazy. We use software and technology to make life easier – and it isn’t always welcome news when that same technology causes us to do a little work that we never intended to do in the first place.That’s why so many Webmasters use tools like Google Analytics, and then move on to other things with the assumption that they have filled the need for a good Web metrics tool. After all, it has the name GOOGLE right in the title – and who understands the Internet better than Google?
To be clear, my intention here isn’t to bash Google Analytics, but rather to warn Webmasters that if you’re using Google Analytics for your sole source of statistics about your website’s performance, you’re almost certainly getting only a partial view of the real picture.
There are a number of reasons why traffic that you receive from Google Analytics might not be 100% accurate. But the purpose of this article is less about why Google Analytics isn’t 100% accurate, and more about fighting the common misconception that it is in fact completely accurate for counting unique visits.
Rather than reinvent the wheel here, I’ve pulled together a collection of snippets and links from around the Internet that will give you an idea of what various experts as well as average Webmasters think about the accuracy of Google Analytics. So without any further ado, here are some comments and links that will provide you with some evidence that you need to take your Google Analytics stats with that proverbial grain of salt.
Let’s start with the entry on Google Analytics from Wikipedia.org, which includes these three paragraphs on its limitations:
“Many ad filtering programs and extensions (such as Firefox’s Adblock and NoScript) can block the GATC. This prevents some traffic and users from being tracked, and leads to holes in the collected data. Also, privacy networks like Tor will mask the user’s actual location and present inaccurate geographical data. Some users do not have Javascript-enabled/capable browsers or turn this feature off. However, these limitations are considered small – affecting only a small percentage of visits.”
“The largest potential impact on data accuracy comes from users deleting or blocking Google Analytics cookies. Without cookies being set, GA cannot collect data. Any individual web user can block or delete cookies resulting in the data loss of those visits for GA users. The only protection a website owner can use to prevent this, is to ensure best practice policies are upheld on their web site. That includes being transparent in what visitor data is collected and how it is used. This information is usually placed within a privacy policy statement page.”
“Because GA uses a page tagging technique to collect visitor information via a combination of JavaScript and cookies, it has limitations with websites browsed from mobile phones. This is due to the fact that only the latest phones are currently able to run JavaScript or set cookies (Smart phones and PDAs).”
http://en.wikipedia.org/wiki/Google_Analytics
More thoughts from blogger Michael Martinez at SEOmoz.org, including some numbers from his internal tests:
“The discrepancies vary across the selection of pages. For Xenite’s home page, we show over 12,000 views in January but Google shows 2200. In July, Google thinks we only received about 800 page views on the main page. Our server logs indicate 3,000. “
“Our server logs indicate we had 97,000 visitors in January and 81,000 visitors in July. So while a greater proportion of our summer traffic comes from non-Google sources, we didn’t lose almost 50,000 visitors — which the Google Analytics data suggests we should have.”
From user comments from that same blog entry:
“No argument here – I’ve been using Analytics since it came out, and it has consistently reported between a quarter and a half of the traffic coming from my logs. It’s frustrating; I like the way you can work with the data in Analytics, but I’m not at all happy with the data it gives you to work with.”
“I have put all my eggs in the Google Analytics basket as I was sure this was top notch software [sic], but at the same time I have found it strange that it underreports so much.”
http://www.seomoz.org/blog/how-reliable-is-google-analytics
From blogger Collin at FeedFlare.ca:
“Something was not adding up when I saw the web analytics report for these two new sites of mine. One of the sites according to Google web analytics had only five visitors, but when I checked where that traffic came from it said that I had 15 uniques from my social bookmarking site called Link Adore plus some search engine traffic.I had even bought traffic once in June to come to this site as a test and seen that Google didn’t pick up any of that traffic but my Awstats recorded all of it and the right amount as well.”
Here’s one of the comments from a reader:
“It is funny that you mention that because I have been thinking the same thing. I do quite a bit of blogging and on some of my blogs I have a little widget called FeedJit that sits on the sidebar. This widget tracks live stats. There is a little chime that will sound off when I get a visitor and at times that chime goes off non-stop. Yet next day when Google Analytics catches up it doesn’t even report half the traffic the FeedJit widget reported.”
http://www.feedflare.ca/web-analytics/how-accurate-is-google-web-analytics
From blogger Tad Chef at SEOptimise.com:
“Anyways, I use Google Analytics myself and noticed before that some numbers are inaccurate or downright wrong. In May I discovered how wrong, completely wrong.”
Example user comment:
“Since last month I am also facing this problem that Google Analytics showing wrong data i checked my conversions in backend there is 12 conversion in a day and same day i checked it in GA i found 0 conversions.”
http://www.seoptimise.com/blog/2008/06/is-google-analytics-cheating-you-use-alternatives.html
From blogger Abhishek Bhardwaj:
“On analyzing the data over this period of time Google showed “679” conversions for my Lead Capturing goal. This meant that the “thank-you” was hit 679 times as per Google Analytics. But to my surprise the number of records in my database was “792”. I could realize the reasons behind this difference to be the JavaScript used in the analytics code (as many of us believe) or users’ behavior which intends them to exit the browser window before the “thank-you” page gets fully loaded.”
He continues…
“I still felt that the JavaScript problem was not a concrete reason for such a big drop-out in the numbers. So, for the second month I planted the Google analytics script in the action event of the submit button. So this meant that whenever someone hits the submit button on successful completion, a goal is accomplished and this increments the number of tracked goals by one. Again the result was as spoofy as the previous one. The tracked goal conversions numbered at 1160 and the actual stored records were 1367.”
http://www.evancarmichael.com/SEO/1246/Google-Analytics–How-Accurate-is-it.html
I’ll stress again, in no way am I suggesting that you should not use Google Analytics, or that the service isn’t useful. The information that it does report is probably very accurate information – it just isn’t 100% complete. In a way, that’s a good thing because it means you can have confidence at least in the traffic you do see reported, and that your actual complete numbers are almost certainly better, not worse.
One more link, if you’re still not convinced: