The 7 Most Common Configuration Problems on Google Analytics (… and How to Fix Them)

1. The classic version of Google Analytics doesn’t count everything

Google initiated the long process of updating Analytics in 2013, and the migration to the new Universal Analytics is complete. This new version provides new ways to collect information, new reporting tools, and includes updated mobile development kits.

Many companies are still using statistics from the classic version, which can no longer guarantee the most accurate data. Here are two ways to find out if your website uses the Universal version:

  • In your website’s source code: If the tracking code mentions “analytics.js” instead of “ga.js” you’re using Universal Analytics.
  • In your Google Analytics account: Go the the Admin > Properties tab. If you see “Tracking information” with the new settings “Session Settings” and “Organic search results” rather than “Tracking code,” you’re using Universal Analytics.

2. Robots and browsers are counted in your statistics

It’s not always easy to identify bots, which can seriously manipulate your web traffic. Incapsula, an online security company, found that over half of a website’s traffic came from either benevolent bots, like search engines, or malicious bots, like spammers. This proportion can reach 80% for smaller websites! To give you an idea of its scale in 2014:

At the request of many webmasters, Google Analytics can now distinguish real traffic from total traffic.
To enable this feature, go to the Admin > Account > Views> View Settings tab, and check “Exclude all hits from known bots and spiders.” You can also improve the fidelity of your results by look at your analytics to identify “non-human elements.” Filter them out under the Admin > Filters tab.

In his blog, David Cole explained how he found robots, for example, by identifying a 100% bounce rate, 0 second visits, etc., and then how he filters them out—in this case by excluding a dubious Mozilla Compatible Agent browser.
Also, look out for spam referrers that have “social-buttons” in their URLs. Ideally, you should take the time to really look through your website requests to find patterns to exclude potential bots. For example, we’ve found that Russia is the number one exporter of analytics spam bots.

3. Your employees count, but their IP addresses shouldn’t

Your employees are definitely throwing your statistics off! One example: Corelia was trying to find out what city our visitors came from. As it turned out, our managers’ towns sat at the top of the list, probably visiting the website to work from home. Funny as it may be, it nevertheless skewed our geographical statistics.

To exclude internal IP addresses, you’ll need to create an exclusion filter:

  • Filter type: Custom > Exclude
  • Filter field: IP Address
  • Filter Pattern: Enter the IP address with a regular expression (ex: 176.168.1.1.)

4. You might know the unknowable

Some reports will have a very unhelpful “not set” line. While we may not be omniscient, we are sometimes able to refine or even to correlate them in certain reports.

AdWords reports Behavior reports (Site Content > Landing Pages) Custom reports
Problem:

The website or app you’re tracking in the Analytics account is receiving traffic from an AdWords account that is not linked to the reporting view.

Problem: A session that doesn’t include a page or a screen view, but does include another kind of interaction hit type (e.g. an events or ecommerce hit type). Problem: When you select a traffic-source related dimension (e.g. Campaign) in a Custom Report, the dimension is applied to all sessions in the account, not just paid search (often referred to as CPC) traffic. Therefore, sessions that do not have any campaign information associated (e.g., direct, referral traffic) will be reported under (not set). You will see something similar if you select the All Campaigns report under Acquisition rather than the Campaigns report under Acquisition > AdWords.
Check that:

  • Auto-tagging is on but cost data is not applied
  • There is a redirect in the URL
  • The gclid parameter is altered or dropped from the ad (See: CPC Data Not Collected)
  • Manually tagged URLs are missing a parameter
Use:

The advanced filter (found at the top of the data table) to restrict the data to include page/screen views matching exactly 0 in the following reports:

  • Behavior > Events,
  • Acquisition > Social > Plugins
  • Conversions > Ecommerce.
Use:

The advanced segment could be more useful than a custom report.

Go to New Custom Report > Audience overview > + Add Segment

Source: The Google Analytics Help Center outlines the possible causes and defines the different fields here.

5. JavaScript codes may lead to conflicts

Javascript could lead to conflicts between Google’s tools and your website, or even between Google’s very own tools. For example, Google Universal Analytics and Google Tag Manager both use the same code to collect data. Having both codes installed means twice the data collection and, thus, inaccurate data.

Some warnings from Optimisation Conversion on Google Tag Manager:

“Using the document.write Javascript method can impact page-load times. Custom scripts might be buggy or conflict with another script on the site, and potentially have a serious impact on user experience. Another example, a third party script could directly, or indirectly (by calling an external script), collect more information than desired…”

For the sake of Q.A., it’s always important to test your Google Tag Manager or Google Analytics (simple) codes at the end of the project to avoid breaking any other JavaScript. It’s rare, but it does happen.

6. Your different Google accounts are better off linked together

For starters, it makes a lot of sense to link your Google AdWords account to your Google Analytics accounts, rather than looking at them independently. This allows you to better measure your ad performance and investments.

To link the accounts, from Google AdWords:

  1. In Tools > Google Analytics, you’re redirected to the Admin tab in Google Analytics.
  2. Select the right account in the “Account” column.
  3. Select the corresponding Google Analytics account in the Properties column and click “Link with AdWords” beneath it.
  4. Select the AdWords accounts you’d like to link with this Analytics property.
  5. In the “Set up link” section, provide a title to identify your group of AdWords accounts.
  6. Approve the link with the last button.

Also, Google Webmaster Tools is a management tool for search engine optimization and it too can be linked to your Google Analytics account, once you’re approved and an administrator. To link these accounts:

  • From Google Webmaster Tools: Site management > Google Analytics Website (expanding menu) > Select the Google Analytics site to link.
  • From Google Analytics: Acquisition > Search Engine Optimization > Click the button to set up Search Console data sharing.

7. It’s essential to set clear goals for the website

What’s the point of writing that email blast to drive people to my website? What visits were generated on product demo pages, or requests for estimates, and how can I increase conversions? When you don’t set goals, you’re expecting nothing concrete from your website! Take a minute to sit down and think about what you need to know for your business and sales strategies.

How to set these goals could be an article in itself! In the mean time, check out some Destination Goal examples from Google Analytics Help.

Need help configuring your Google Analytics account? Get in touch!

Sources:

Google Analytics Help Center:
https://support.google.com/analytics/?hl=fr#topic=3544906

29 Common Google Analytics Data Errors And How To Fix Them
https://blog.kissmetrics.com/google-analytics-data-errors/

Why Most Marketers Fail at Analytics
https://www.dynamicyield.com/2015/02/web-analytics-fail/

Why Your Google Analytics Bounce Rate is Wrong
http://blog.popcornmetrics.com/why-your-google-analytics-bounce-rate-is-wrong-and-how-to-fix-it/

Common Google Universal Analytics Mistakes that kill your Analysis & Conversions
http://www.optimizesmart.com/common-google-analytics-mistakes-that-kill-your-analysis-reporting-and-conversions/

Astuce : Identifier (et bloquer !) les robots d’indexation dans Google Analytics
http://davidcarlehq.com/identifier-robots-indexation-google-analytics/

Google Tag Manager – meilleures pratiques pour gérer GTM
http://www.optimisation-conversion.com/webanalytics/google-tag-manager-meilleures-pratiques