Google Analytics Health Check: Is Your Configuration Broken?

Google Analytics Health Check

Peter Drucker famously said, “What gets measured gets managed.” But what if your measurement data is incorrect? What if you’re not measuring correctly or completely? What if there’s a whole pile of things you think you’re measuring when really… you’re not?

The fact is that a lot of the people relying on Google Analytics are relying on bad data. No, not because Google Analytics is awful. Because their configurations are broken.

And if your configuration is broken? Well, you’re likely managing the wrong things and making poor choices based on incorrect (or incomplete) data.

What’s a Google Analytics Health Check?

In the past, we’ve covered setting up and using Google Analytics fairly extensively. If you haven’t already, you can read our Google Analytics 101 and Google Analytics 102 guides. More recently, we covered segmentation and how to do it right.

Still, things go wrong. Problems arise. Errors come up.

A Google Analytics Health Check is a series of checks that help you answer the following three questions:

  1. Am I collecting all of the data I need?
  2. Can I trust the data I’m collecting?
  3. Is anything broken or tracking / reporting incorrectly? Why?

A checklist will guide you, but this is an exploratory mission.

If you’re an agency or freelancer, this is especially true. Since you might not have had control over the Google Analytics setup stage, you might not know what to expect. Unexpected or irregular issues might arise that you hadn’t even heard of. As time goes on, add these types of issues to your checklist.

Step 1: Property Settings

Property Settings

Within the Property administration panel, here are the questions you should be asking yourself…

  • Is the default URL set up correctly?
  • Are your referral exclusion settings configured correctly (e.g. you use PayPal to process payments)?
  • Is enhanced link attribution turned on?
  • Have you enabled demographics and interest reports?
  • Is Google Webmaster Tools linked properly?
  • AdWords Integration
    • Is it configured correctly?
    • Is PPC data showing in Google Analytics?
    • Are the resulting clicks and sessions being recorded properly?

Step 2: View Settings

View Settings

Within the View administration panel, here are the questions you should be asking yourself…

  • Are your Views set up correctly?
  • Is eCommerce tracking turned on?
  • Is site search tracking turned on?
  • Do you have your “Virgin View” and “Working Views” configured properly?
  • Are you doing country filtering?
  • Is your default page and time zone configured properly?
  • Filter Creation
    • Is the office IP address filtered out?
    • Are the IP addresses of your affiliates filtered out (agencies, freelancers, etc.)?
    • Is your home IP address filtered out?
    • Are the IP addresses of remote employees filtered out?
  • Are your Goals set and tracking properly?
  • Have you configured your custom and default channel groupings?

Step 3: Common Issues

1. AdWords Account Not Connected

  • What It Is: You can sync your AdWords account to your Google Analytics account to monitor impressions, clicks and cost.
  • Solution: Admin > Property > Product Linking > AdWords Linking.

2. AdWords Auto-Tagging Not Used

  • What It Is: Auto-Tagging adds a unique ID to the end of the destination URL. Using this, you can calculate your advertising ROI by combining that data (campaign, keyword, cost per click) with Google Analytics Goals.
  • Solution: Admin > Property > Product Linking > AdWords Linking.

3. Time Zones Not Matched

  • What It Is: Your Google Analytics and Google AdWords accounts should be in the time zone your campaign targets.
  • Solution: Admin > Choose a View > View Settings > Time zone country or territory.

4. PPC Keyword ID’s Visible on Landing Pages

  • What It Is: When PPC keyword ID’s are visible on landing pages, they are harder to analyze because Google Analytics thinks they’re two different pages.
  • Solution: Admin > View Settings > Exclude URL Query Parameters (e.g. OVKEY or OVRAW).

5. Bing Tagging Not Set

  • What It Is: If you don’t tag Yahoo and MSN paid traffic as Bing traffic, the traffic will be recorded as organic.
  • Solution: Ensure your Bing adCenter landing pages are using a tag that includes ?utm_medium=cpc&utm_source=bing.

6. Manual UTM Tags Not Used

  • What It Is: Adding manual UTM tags ensures your channels are being properly credited. Offline marketing, newsletters, social media, affiliate links, etc. all need these tags.
  • Solution: Use the Google Analytics URL Builder.

7. Site Search & Category Search Not Enabled

  • What It Is: Site search and category search show user intent, and provide additional keyword lists for PPC and search campaigns.
  • Solution: View Settings > Set Site Search Tracking to ON > Enter the parameter(s) that indicate(s) a site search > Enable categories > Enter the parameter(s) that indicate(s) a category search > Apply.

8. Not Tracking “mailto”

  • What It Is: You can easily track when (and which) email links are selected on your site.
  • Solution: Use an Event push in your email links.

9. Homepage Filter Not Used

  • What It Is: If you don’t filter your homepage correctly, you could end up tracking multiple homepages (e.g. / and /en/ and /fr/).
  • Solution: Admin > All Filters > + New Filter.

10. Log Spam Not Filtered Out

  • What It Is: If you don’t add a filter to remove staging traffic or traffic from developers, you’ll be acting on false data.
  • Solution: Admin > All Filters > + New Filter > Filter this traffic out by hostname.

11. Error Pages Not Tracked

  • What It Is: Broken pages are low hanging fruit, if they can be identified easily.
  • Solution: Add your Google Analytics code to 404 and 500 pages.

12. Duplicate eCommerce Data Not Filtered

  • What It Is: Duplicate data occurs when a visitor (a) refreshes the page, (b) uses the “Back” button, restores tabs from a closed browser, etc.
  • Solution: Use a transaction ID and time stamp when tracking eCommerce transactions. If you’re especially tech-savvy, you can handle this server-side.

13. Bounce Rate of Less Than 10%

  • What It Is: Start by looking at pages where entrances are greater than 100 and sort them by bounce rate. If your bounce rate is less than 10%, something has gone wrong.
  • Solution: Ensure you haven’t added two scripts to the page. If not, visit the site and look for other issues that could be impacting your data.

Every good conversion optimizer has a Google Analytics health checklist of some sort. Here’s a look at part of Peep’s checklist…

Peep LajaPeep Laja, ConversionXL:

  • “Missing tracking code on some pages, resulting in self-referrals or visits not recorded.
  • Missing tracking code on 404 pages and server 500 pages.
  • setDomainName missing on blog.mydomain.com, resulting in referring keyword lost and 2 visitor sessions.
  • set commands, such as Custom Dimensions.
  • iFrame banner tracking resulting in double cookie.
  • Event or double pageviews for same account being called onload, resulting in 0% bounce rate.
  • CustomVar called after pageview.
  • Missing eCommerce category.
  • sessionID’s in URLs.
  • Thank you or sale complete pages not using separate URLs.
  • Site Search using static URL or mod-rewriten URL, thus q=keyword not extractable.
  • Quotation marks not escaped or ansi characters not encoded in eCommerce category or item names.
  • Server redirect is stripping off gclid.
  • Meta refresh redirect causing gclid to be lost.
  • Missing manual tags in emails and newsletters & RSS, social, Google products, Google News.
  • Inconsistent tracking used; mixing ga asnc and legacy code.
  • Full URL entered in Goals or Goal matching too broad due to “header setting”.
  • Blending of goal values + transaction values.”

Annie Cushing of Annielytics checks Google Analytics health using an audit…

Annielytics Audit

You can get access to the full audit spreadsheet here. For more advanced help, she also has a self-guided site audit template you can purchase.

For now, here’s a look at some of the items on her free checklist…

“AnnieAnnie Cushing, Annielytics:

  • “Has the site had any significant drops in organic traffic?
  • If the site has experienced drops, are they seasonal?
  • If the site has experienced drops, do they correlate with any major algorithm changes?
  • What tools are they using for tracking?
  • Are they tracking conversions?
  • Are they using ecommerce tracking? (See more checks at the bottom of the list.)
  • Is their analytics tracking code missing from any pages?
  • Do other sites have their GA code on them?
  • Do they have subdomains?
  • If the site has subdomains, does GATC include _setDomainName() method?
  • If the site has subdomains, are they including hostname in content reports?
  • If so, is GATC set up properly?
  • Do they have PPC campaigns showing up in organic results
  • If the site offers site search, are they tracking it in their analytics?
  • If so, does their site search appear to be effective?
  • Are they using asynch?
  • Is site is using asynch, are pages on the site still using traditional?
  • Are they using annotations?
  • Have they set their homepage in GA?
  • If running Google AdWords campaigns, is your AdWords account linked w/ GA?
  • If running Google AdSense campaigns, is your AdSense account linked w/ GA?
  • Do content reports contain utm parameters?
  • Does the site use a third-party cart? If so, do they have cross-domain tracking in place?
  • Does the profile use filters correctly?
  • Should the domain use URL rewriting?
  • Does the site have excessive sampling? If so, is it b/c they have more than one website in a property?
  • Does site use campaign parameters on internal links?
  • Do any pages use meta refresh? (This will artificially lower bounce rate.)
  • If the site runs email campaigns, are they tagging URLs that point back to their site with campaign parameters?
  • Does the profile use event tracking? If so, are they naming category, action, and label as they’re intended?
  • If they’re using event tracking, are they using events as goals?
  • Does the site use custom variables? If so, are the being scoped properly?
  • Do custom variables share the same slot?
  • If site uses ecommerce tracking, is the tracking code on all conversion pages?
  • If site uses ecommerce tracking, are there JavaScript or server-side programming errors before the _trackTrans() method keeping it from firing?
  • If site uses ecommerce tracking, does the code have currency symbols or or thousands separators in code?
  • If site uses ecommerce tracking, do their products or store ID use apostrophes?
  • If site uses ecommerce tracking, is it international / use multiple currencies?
  • Does the site have a YouTube channel? If so, does it track it in GA?”

If you don’t have the time (or interest) to perform your own site audit, you can use an app.

4 Common Tracking Issues

1. The Basics

Starting with the basics, here’s a list of common tracking issues that Google has publicly published…

  • Using incorrect snippet and/or viewing the wrong account or view
    If you track multiple websites and/or have access to multiple Analytics accounts, you might be using the snippet from another account and/or view. Make sure you are viewing the correct account and view. See Finding the Tracking Code for additional information.
  • Extra whitespace or characters
    Be sure to copy the snippet and paste it directly onto your website using either a text editor or an editor that preserves code formatting. Don’t use a word processor to copy the snippet from your account. Doing so can add an extra space or change the quotation marks in the tracking snippet, which requires precise formatting in order to work.
  • Customization errors
    If you are making customizations to the tracking code, make note of the following:

    • function names are case sensitive and should have correct casing
    • boolean values (e.g. true or false) should not be enclosed in quotes
  • Incorrect filter settings
    Incorrect filter settings can affect the data you see, and can inadvertently filter all of your data from your reports. In most cases, this occurs when users apply multiple “Include” filters. For details, see the article on Include and exclude filters.
  • Other scripts on your page
    If you’re running other scripts on your page, make sure you’re not using any variables that Google Analytics uses. For more information on variables that the Google Analytics tracking code uses, visit our Developer Guides for the Universal Analytics (analytics.js) JavaScript library, and the Classic Analytics (ga.js) library.

2. Missing Pages

To identify pages that are missing your Google Analytics code, you can look for irregularities in your data. Or, you can use a tool like Google Analytics Checker.

Once you’re sure every page of your site has the code, you need to ensure it’s the latest (asynchronous) code.

This means that instead of Google Analytics loading synchronously, it will load asynchronously to avoid blocking resources that load later on the page. Essentially, it enhances the speed that the tracking code is loaded.

Click here for more information on asynch.

3. Different Data in Your Shopping Cart Tool

If you’re in the eCommerce space, you’re likely using some sort of shopping cart tool. So, what happens when the data showing in your shopping cart tool is different than the data showing in Google Analytics?

Essentially, there are four possible issues…

  • Your Google Analytics eCommerce Tracking is not properly installed. Read this thorough guide to setting up eCommerce Tracking to ensure you’ve done it correctly.
  • Time zone. If your shopping cart tool and Google Analytics are configured to report in different time zones, you may have mismatched data.
  • Time of day. If you set up your eCommerce Tracking in the middle of the day, transactions that occurred before will not appear in Google Analytics, but will of course appear in your shopping cart tool.
  • Cancelled transactions. Transactions with no value ($0) and cancelled transactions do not appear in Google Analytics.

4. Cross-Domain Tracking

You’ve seen cross-domain tracking mentioned a few times now. What is it exactly? Chris Mercer from SeriouslySimpleMarketing.com explains…

“ChrisChris Mercer, SeriouslySimpleMarketing.com:

“It’s called ‘Cross-Domain Tracking’ and it could come into play if your client’s site has multiple domains as part of their funnel or ‘buyer’s journey’. In these cases, you’ll absolutely want to setup Cross-Domain Tracking.

I know what you’re thinking… ‘How do I do that?’

If you’re using the traditional Google Analytics (boring) try doing this.

If you’re using Google Tag Manager (you should be) it’s faster to do this.”

For example, your checkout process might be on a different domain. Unfortunately, Google Analytics uses first party cookies, which can only be read by the domain that issued them. So, in order to do cross-domain tracking, you need to share cookie information with the different domains involved.

An alternative to the Google Analytics resource Chris provides above is Optimize Smart’s Google Analytics Cross Domain Tracking – Complete Guide.

3 Common Google Tag Manager Issues

Now, there’s a good chance that some of you are using Google Tag Manager. Of course, there are common tracking issues associated with it as well. Here are the top three…

  1. Tag Isn’t Firing – There are a number of reasons your tag might not be firing. You have unpublished changes, your triggers are too specific, your triggers are configured incorrectly, etc. Find a full list and begin troubleshooting here.
  2. Wrong Filter Settings – When you apply multiple include Filters, you can accidentally end up filtering all of your data out of your reports. Read up on how to properly use include Filters (the hit is discarded if the pattern does not match the data) and exclude Filters (the hit is discarded if the pattern does match the data).
  3. Unpublished Container – Before adding the tag, be sure you published the container. Changes you make to a container will not be saved until you publish the container. For more information on publishing containers, click here.

Conclusion

Without a Google Analytics Health Check, you’re analyzing data with both of your eyes closed. [Tweet It!]

Don’t be one of the many people making important business decisions based on bad data from a broken Google Analytics configuration.

Here’s what you can do to get your Google Analytics back in shape…

  1. Create a Google Analytics Health checklist that you can use to evaluate your setup.
  2. Go through your Account, Property, and View settings and ask yourself the questions above.
  3. Review the 13 most common Google Analytics issues and check to ensure you’re not making them.
  4. Review the 4 most common tracking issues and check to ensure you’re not making them.
  5. Review the 3 most common Google Tag Manager issues and check to ensure you’re not making them.
  6. Run through your checklist annually. As time goes on, new issues can arise.

Join the Conversation Add Your Comment

  1. With the title, I expected this to be another basic Google Analytics article. The tips from Annie and Mercer alone are going to save me thousands of dollars. Thanks for linking to that spreadsheet Shanelle!

    1. Shanelle Mullin

      Happy to help, Luiz! Glad you found it helpful and it wasn’t another basic Google Analytics article.

  2. The need to filter out spam refferals canot be overstated. This rticles touches upon it but Google it for indepth solutions.

    Spam referrals can play havoc especially with low traffic sites. Bounce rates can go way up for example. I’ve seen bounce rates increase by 50% due to spam referrals. Thanks for the tips.

    1. Shanelle Mullin

      Thanks for reading and for sharing the additional tips, Jacob! Glad you liked the article.

  3. I’m going to certainly bookmark this page, and going to pass the link to freinds and business owners.

    Such a good check-list without burning the money upfront.

    One more thing is that, we need to modify the conversion tracking. So that we can track the total amount of the cart with shipping etc.

    You should include that as well :)

    Thanks for the list.
    Cheers
    Suresh

    1. Shanelle Mullin

      Thanks Suresh! Glad you found the article helpful overall.

      The lists from Annie and Peep are awesome.

Comments are closed.

Current article:

Google Analytics Health Check: Is Your Configuration Broken?