While Google Analytics is one of the best free analytics applications out there for getting basic traffic and visitor data, it also has some pretty significant holes. You’ve probably noticed some discrepancies when comparing your Google Analytics data to other analytics sources which can include anything from your hosting company’s own built in analytics to social analytics programs that track hits from shortened URLs sent through Twitter, Facebook, or other networks.
There are many reasons for confusing and missing data. Here are just a few examples of the areas from which you might be getting conflicting data and the reasons why.
Internal Traffic
If you are just one lone person operating a website that you visit maybe once a day, this may not be a big issue to contend with. But if you are someone who looks at your site multiple times a day, or have a company whose employees are viewing the website multiple times a day, then one thing Google Analytics might be hiding from you is the fact that your visitor statistics, pageviews, and other data may include you, your webmaster, your employees, and other people you don’t need to analyze. Just imagine how much your bounce rate is if you are testing a page repeatedly by opening, refreshing, and closing it immediately after?
To exclude one IP address in Google Analytics 5, you can go to your website profile settings, Filters, and click on +New Filter. Name your filter appropriately, then use the dropdowns to Exclude traffic from the IP addresses that are equal to, and enter your IP address.
If you’re not sure what your IP address is, you can actually search Google using the query what is my ip address and it will tell you right at the top of search results.
If you need to exclude a range of IP addresses (like you would with an office), you can use the IP Address Range Tool to enter the first and last IP address in the range that your office uses, and it will generate the expression you will need to use in your exclusion filter.
You can learn more about excluding internal traffic filters by IP in Google Analytics Help.
Direct Traffic
The next missing piece of data in Google Analytics is direct traffic which can be found under Traffic Sources > Sources > All Traffic and is displayed as (direct) / (none) This traffic is essentially any traffic that comes from an unidentifiable referral source. It’s not from social media, search engines, or other referring URLs.
Direct traffic includes people who have:
- Typed your URL into their browser’s address bar.
- Your website bookmarked on their browser’s bookmarks bar.
- Clicked on links in documents or emails.
- Come from sites that do not create a live hyperlink to your website, but rather just have an inactive link that has to be copied and pasted.
- Security settings preventing referring URL’s to pass through to another website.
The upside is you still have the rest of their behavior tracked in Google Analytics once they have reached your site up until the point they leave. You just don’t get to find out where they have come from. You can learn more about traffic types, including direct traffic in Google Analytics Help.
A few things interesting to note about direct traffic is the percentage of new visits (people who have bookmarks on their browser are probably repeat visitors) and related data which you can find by going to Traffic Sources > Sources > Direct. I have found traffic under the direct traffic area have a lower bounce rate, higher average time on site, and higher rate of goal completions.
Mobile Traffic
Google Analytics does give you data from mobile devices, but not everything. You can view various stats such as your mobile vs. non-mobile visitors.
You can also see specific mobile device info about your visitors.
Of course, you can see that you’re not getting the complete picture with the large number of (not set) devices. When you add in the screen resolution and sort, you can see that the issue possibly lies in Google Analytics not being able to track the latest iOS for Apple devices.
Essentially, Google Analytics can only track so much from mobile phones, and traffic from mobile applications may not be included. You can learn more about mobile reporting availability in Google Analytics Help.
Private Browsing Traffic
Popular browsers now offer users the option of private browsing. This means that their activity on your website will not be tracked in Google Analytics at all.
Google Analytics Opt-out Browser Add-on is provided by Google itself and is available for all of the top browsers including Internet Explorer, Chrome, Firefox, Safari, and Opera. There are also many extensions / add-ons made for specific browsers that blog all types of analytics software tracking, so no matter what you use, you may still be missing some data!
Keyword Traffic
The most recent change in Google Analytics has everyone stirring in the online marketing world is the loss of keyword data from organic search. When you look under Traffic Sources > Sources > Search > Organic, you might see that your top referring keyword is now (not provided).
This recent change is the result of Google changing the way logged in users search. When you search Google while logged into your Google account, you will be searching in SSL mode. Searches in SSL mode allow logged in users to have more privacy when they search by not allowing webmasters to see the keywords that these users searched to arrive to their website in organic search results.
Although the change was made for the sake of privacy, most believe that is not the case as logged in users will still have their keywords tracked and sent to webmasters who use Google AdWords. In this case, the keywords that led to a logged in user clicking on a paid listing will show in Google Analytics.
Regardless of whether the data can be accessed by webmasters in Google Analytics, the data is still being tracked. If you are logged into your Google account right now, you can visit your web history to see what is still being collected about your search behavior by Google.
Even if you shut this off, Google will still collect this data.
Social Media
Last, but not least, is the inconsistencies when it comes to social media tracking in Google Analytics. I firmly believe you can learn a lot through Google Analytics through the use of advanced segments to see what drives traffic to your website in terms of social media vs. other referral websites.
But if you start to compare other analytics data for social media to that of Google Analytics, you might be in for a surprise. Several applications and tools such as Bit.ly, HootSuite, and Buffer will show you click data for links you share on Twitter or other social networks through their platform.
Yet, if you look at your Google Analytics, these numbers may not match up. Again, this can go back to any of the previously mentioned items such as mobile application traffic, direct traffic, or people using private browsing.
Do you know of other pieces of data that Google Analytics is hiding from you? Please share it in the comments as well as the ways or alternatives you use to find out more about this missing data.
About the Author: Kristi Hines is a freelance writer, blogger, and social media enthusiast. Her blog Kikolani focuses on blog marketing, including social networking strategies and blogging tips.
Awesome Infographics ♥ Marketing Guides ♥ User Feedback Software ♥ Customer Analytics Software
Fantastic article! I was just in the middle of trying to work out my Google Analytics and came across your Twit – Thanks for the help
You’re welcome Maria! :)
helpful article , thanks a lot.
Great article. One question. It seems like iPhone screen resolution is shown in Google Analytics as 320×480. This is true for the older iPhone, but not as of iPhone 4 (640×960). Why is 640×960 not listed?
Not sure about that Martin. I was just going off of my own analytics, and so far I don’t see that set of dimensions. I did read that they were having issues tracking the new ios, so that may be a contributing factor.
“…direct traffic area have a lower bounce rate, higher average time on site, and higher rate of goal completions.”
This makes a lot of sense, and matches my experience with websites I have bookmarked.
KISSMetrics should just hire you now while they can. Emeryville is very nice, nestled on San Francisco Bay. You could live in The City, reverse commute is a snap!
Thanks Dave, although I’m pretty attached to Arizona and working from home for the time being. :)
Thanks Kristi for the post.
Your first point regarding internal traffic is to the point & I must adhere that I too think that google hides this data. In my company there are several developers & they are working full 9 hrs on an ecommerce site testing, developing, adding new features etc.but after adding the IP address as you mention I normally get several visits from my own company which I don’t understand the purpose of google for providing this exclude your internal traffic filter.
Direct traffic points are correct, except of clicking on links in document or emails, I was thinking tha that will come under referral. Can you explain on that?
How can you be sure that it is the traffic from latest iOS 5 Apple device?
Well, for private browsing, I think that Google will hide that results even if you bribe them to show it. Like, if I’m searching something private (without logged in) then why do I want that keywords to be shown to third party analytics software, isn’t that?
And for SSL Encryption, the same applies as above & also stated by matt cutts in pubcon that “if you search anything private with logged in those terms will not shown in analytics referral traffic” here I’ve to differ with him because if I want to search anything private I’ll without logged in.
Great Post! Still trying to grasp whether the change Google made in October in tracking keywords can account for so many keywords appearing as (not set) in our PPC campaigns when viewed in analytics. From your article, and google’s initial blog post, it appears the change should only impact organic searches.
Strange, it was only supposed to affect organic search. That’s why everyone was up in arms over it – Google said it was for privacy, yet if you pay for AdWords, you could get past the privacy barrier.
Good post!
But what are we supposed to do about this? Yes, there are flaws, but how are we supposed to address them? Is there some other visitor analysis software that works better?
Okay…here’s my question. Do you trust bitly stats over Google Analytics? I see there were problems a couple of years ago with bitly tracking spider hits, etc or having inflated numbers.
Would you trust bitly numbers now over GA?
Along with this point, I have seen massive click spikes due to bots/spiders. You can test this yourself by creating a Twitter account with a random string of characters. No follows or followers, and tweet out a shortened link that is also randomized (no vanity). Last time I tried this I got 15-20 registered “clicks” on the link, that were clearly not human clicks.
It would only follow that any additional shares or “RTs” would cause this number to artifically rise further. The click stats are interesting to track in relation to each other, but don’t aren’t a valid measure of actual user behavior.