Are Google Analytics accurate?

There are many different sources for web analytics. You can set up free Google Analytics on your site, you might get analytics from your website or blog host, and you might have paid for an advanced analytics tool. But what happens when they all return different numbers in their reports? Which one is right?

There are many different sources for web analytics. You can set up free Google Analytics on your site, you might get analytics from your website or blog host, and you might have paid for an advanced analytics tool.
But what happens when they all return different numbers in their reports? Which one is right?

google-analytics-accurate

We are not our devices

How accurate are google analyticsFirst of all, nothing is 100% accurate and precise. None of the analytics packages tracks exactly how many humans are visiting your site, reading your pages, and clicking on your links. Analytics tools track visits from computers, not necessarily distinguishing between who is using the computer to access your site. Different people can use the same computer, and a single person can use different devices to access your site. And if a visitor takes a phone call while browsing your site, the time they spend on a particular page can be miscalculated.

JavaScript, Cookies, and Bots (oh my)

Different analytics packages also use different technologies to count visitors to your site, which can lead to discrepancies in the data. Google relies on JavaScript tags to track visitors. Since around 10% of users have JavaScript disabled, you miss out on counting 1 in 10 visitors right away. Google Analytics also only tracks visitors who have cookies enabled on their browser, and cookies get lost, blocked, and deleted over time. No cookies, no count.

Google Analytics counts ignore visits from bots or spiders that crawl your site to fetch content for search engines. However, your web hosting service or blogging platform might count all of these visits. Some analytics tools place an invisible image on each page and whenever that image loads, a visit is counted. Since this count doesn’t rely on JavaScript or cookies, the count might be higher. Even one of Google’s other analytics tools, Google Webmaster Tools, will track visits to pages without the JavaScript tag if users reach them via search results or if Google crawls or otherwise discovers them.

Timing

Google Analytics JavaScript code is usually in the footer of the page, so Google Analytics will not count a visit if the user leaves before the page fully loads. Google Analytics also only counts a visit to the same page if it’s not within the same 30 minutes, weeding out duplicate visits to your site that happen over a short time span. Your other analytics tool might count these visits, leading to a higher visitor count. And there may be a lag between collecting and displaying the results, which can make counts from different sources very different.

Keyword Counts

Reports that tally the keywords that visitors use to find your website are defined differently by different analytics tools. Google Analytics includes counts for search engine queries and paid keywords in the single “keywords” count. Google Webmaster Tools displays only the most significant words Google found on your site. And some tools round the results to one or two significant digits, which can make a big difference in lightly visited sites. And Webmaster Tools shows only search queries from Google, whereas your other tools may combine search results from all search engines.

Are analytics useful?

Each analytics tool tracks the data differently, tallies up the data differently, and reports differently.  If you compare the visitor numbers tracked by Google Analytics to those tracked by your web server, the Google number is almost always a smaller number. But while the numbers aren’t exact, the data is still very useful. You can compare sets of data, look for bigger patterns and trends, and combine data to tell the complete story. You can read more about gaining insights through comparisons in “The Right Way to Use Analytics.”

9 thoughts on “Are Google Analytics accurate?”

  1. Use Google Analytics Counter Tracker plugin on Word Press, it is the best for your blog analysis. You are able a whole lot of functionality just by using the plugin. For example you can Connect your Google Analytics account and get Google Analytics stats;Group stats by hour, day, month and year;See data for different time periods in your reports;Compatible with almost every version of Word-press.

  2. It’s not true that visitors who don’t have JavaScript or cookies enabled aren’t counted in GA. If JS is disabled, your browser will load a utm.gif pixel that will register the visit. Not having cookies enabled will distort the data, but again, you will still be counted.

  3. Yes, it is a huge difference between what Blogger says and what Google says. Best thing you can do is just use the analytics to compare trends, compare between content, or compare what different audiences are doing. There’s just no way to use the numbers as actual visits/visitors with total confidence, alas!

  4. What I dont understand is why google itself gives me two conflicting numbers.

    I use google blogger. If I look at the built in google analytics (via my blogger dashboard), for example, it says I got 1,200 page views this past Saturday.

    If I look at googles separate analytics website…..it says i got 200 page views.

    Thats an enormous difference.

  5. Thanks for the article Kirsti, helped me a lot to get what’s going on.

    I use both Google Analytics and WP’s (old) Site Stats, and of course always end up with totally different numbers.

    I read somewhere that in order to actually compare the numbers, I should rather look at the pagevisits on GA, since that is what is presented on the WordPress stats. Do you know if that’s true?

    Take care:o)

    Shmuel

    1. Google’s definition of “visit” and WordPress.com’s definition of “view” are probably not exactly the same (different proprietary algorithms) but could return similar results. Remember that even such simple things as WordPress displaying your stats by Greenwich Mean Time, while Google uses your specified timezone, can make the numbers differ. Add to that the fact that different WordPress plug-ins might track the data differently and you can see why your analytics programs can return such varied results.

      But, though they return different numbers, the data from either analytics source should show similar trends over time, as well as similar comparisons between content and between segments. See “The right way to use web analytics” at http://www.hotdesign.com/marketing/right-way-to-use-web-analytics/ for more on using the numbers.

      HTH!

  6. I think it’s better to compare the data with different web analytic sources,so that we would get an idea what’s the correct value.

    1. So true, but many of the analytics programs charge for their services so it would be cost-prohibitive to set up all of them. But, in an ideal world, it would be a great idea!

Leave a Reply

Your email address will not be published. Required fields are marked *