7 Google Analytics Issues that Ruin Data Quality (& How to Fix Them)

Clean, reliable data is the foundation of informative marketing analytics. Without a solid technical foundation, companies can be left to make decisions about their website and marketing activity based on spurious findings and false assumptions. A detailed audit of your Google Analytics account is to identifying and fixing common Google Analytics issues.

In this post, (the first of a short series on common Google Analytics configuration issues) I look at 7 of the most common View and Property-level configuration issues affecting Google Analytics, how they impact data quality and how to fix them to improve your Google Analytics configuration and marketing analysis.

Common View Level-issues in Google Analytics

Mistake #1: 'Raw' and 'Test' Views haven't been configured

This is one of most common mistakes I see when auditing a company’s analytics setup. All too often, companies delve into Google Analytics headfirst conducting all their analysis and reporting from the default View, without setting up other critical Views in the process.

Solution: Create additional customised Views

At a bare minimum, every Google Analytics account should contain at least the following Views:

  • Master view (containing IP filters for your own company and any third parties you work with)
  • Raw View (no filtering in place)
  • Test View (to allow experimentation of new Filters, Goals and configurations)

Depending on the nature of your business, it might also be important to set up other business-specific Views (and it likely is). Do staff access your website a lot for company updates or internal communications? Or, maybe you get the majority of your traffic from Mobile, or Referrals play a crucial role in your business? In each of these cases, specific Views should be configured to only show the impact on these traffic sources. Segmented data is so under-utilised, yet so powerful.

To set up a new View in Google Analytics, navigate to the ‘Admin’ area of your account. Click on the ‘View’ drop-down menu on the top-right, and select ‘Create new View’ and name your View appropriately.

Remember, you need to add Filters to Views to truly customise them – it’s best to think of Views and Filters as closely connected in this way (more on Filters next).

Google Analytics audit

Mistake #2: Bot Filtering hasn’t been enabled

You might not even realise you have a spam traffic issue, but if you don’t have Bot Filtering enabled in your View settings, you might well have one.

Why is this a big deal? More often than not spam bots or crawlers will hit your site, ‘view’ one or two pages, then leave. These fake sessions skew the reliability of analytics data and can artificially impact a website’s core metrics like Bounce Rate, Conversion Rate or Pages/Session. Getting a handle on spam by enabling Bot Filtering or Custom Filters is crucial.

Solution: Enable Bot Filtering

To enable Bot Filtering (Google’s own in-built protection against known bots), simply check the following box in your View Settings panel in the Admin area:

Bot filtering

Mistake #3: Spam Filters haven’t been implemented

This should really be done in tandem with implementing Bot Filtering. To check for the presence of spam traffic, use the Referrals report and filter the report to show only high Bounce traffic sources (say 95% or above). Generally speaking, spam bots and crawlers will have an unusually high Bounce Rate.

Also, be sure to look for suspect-sounsing domains like “freetraffic.com”, or “seo-traffic.xyz”. If you have a lot of these sources, your key metrics are going to be affected and these should be added to your Custom Filtsr as below.

Solution: create a Spam Traffic Filter

Filters can be added at either the ‘Account’ or ‘View’ level depending on which best suits your needs. To set up a Custom Filter in Google Analytics:

Select ‘Filters’ in the Admin section

  1. Select ‘Add Filters’
  2. Choose ‘Create New Filter’ and enter a name for your Filter, e.g ‘Spam Filter’
  3. Under Filter Type, select ‘Custom’, select ‘Exclude’ and set the Filter Field to ‘Source’
  4. Paste in any spam Sources identified from your Referrals report
  5. Separate these with the pipe | and include a backslash before any special characters, e.g: free\-fb\-traffic\.com|ecommerce\-seo\.org
  6. Click ‘Save’

Before you hit save, your Filter configuration should look something like this:

Spam FIlter

Mistake #4: Valid hostname Filter not added

This is one that’s often not implemented, but if done correctly can be another effective way to combat spam traffic. Essentially a valid hostname Filter tells Google Analytics “only show me traffic where the hostname is my website”. Now, if you have additional sub domains you need to report on in your analytics View, you’ll want to include these here too.

Solution: set up a valid hostname filter

Implementing the valid hostname filter is incredibly easy. Similarly to the spam filter, follow these instructions:

  1. Select ‘Filters’ in the Admin section
  2. Select ‘Add Filters’
  3. Choose ‘Create New Filter’ and enter a ‘Valid hostname filter’ as the Filter name
  4. Under Filter Type, select ‘Custom’ and ‘Include’ and set the Filter Field to ‘Hostname’
  5. Paste in your domain e.g angus.carbarns.me
  6. Click ‘Save’

Common Property configuration issues

The second set of common issues covered in this post are the common configuration issues that are made at the Property-Level. The main issue with Property-Level mistakes is that any Views under a given Property will inherit its issues, potentially ruining entire datasets. Getting Property-level settings right is key to successful tracking and reporting.

Mistake #5: Misuse of Referral Exclusions List

The Referral Exclusions List is probably the most poorly named feature in Google Analytics. Common sense would suggest this is where you’d filter out unwanted Referral sources like spam and bot traffic. In fact, many people use it for exactly this purpose but this is wrong. Plain and simple.

The Referral Exclusion List plays a role for sites in which users might jump between domains where tracking might not exist, such as between the core website e.g ‘example.com’ and a payment provider or external source, like PayPal.

Why is the Referral Exclusion List important?

Now, in most cases unless that external journey, in this case PayPal, has been added to the Referral Exclusion List, it will show up as a Referral Source in your Referrals report, messing up your marketing attribution. This is because Google Analytics will see sessions leaving the domain, and re-entering from a totally different domain ‘paypal.com’. As such, it’ll regard that as a new session from a Referral source, when we know that’s not the reality.

Solution: using the Referral Exclusion List effectively

If you’re running a fairly simple website, you may not need to use the Referral Exclusion List. However, if you’re running an Ecommerce site that links off to a payment provider (breaking a session), and that provider sends a session back to your site, you will want to add the payment provider domain to your Referral Exclusion List.

By doing so, Google Analytics won’t regard the payment provider as the source of the conversion and your attribution won’t suffer. However, correctly implementing cross-domain traffic is also a crucial step (more on this below).

Mistake #6: Cross-Domain tracking not enabled

If your website contains user journeys that span multiple domains, it’s critical you implement cross-domain tracking. It sounds obvious, but without cross-domain tracking in place, you won’t be able to see journeys between domains, and instead will be left with pages with abnormally high Exit or Bounce Rates that don’t reflect what’s really going on.

Simo Ahava has put together this exhaustive and excellent resource on diagnosing and remedying cross-domain tracking issues which I highly recommend you check out.

Mistake #7: AdWords or Search Console not linked

If you’re advertising through AdWords, or engaged in SEO you should ensure AdWords and Search Console are both linked to your Google Analytics Property and sending data.

Personally, I find the AdWords interface horrendous for conducting analysis or trying to identify additional opportunities, and find it so much more useful to integrate this data with Google Analytics for scaled up reporting and analysis. I find this tends to be the case when analysing Search Console data too, so I prefer to analyse data in Google Analytics or Google Data Studio.

By integrating these other data sources into your main Analytics workflow, you have access to many more reporting options and segments to scale your analysis and tie marketing behaviour to on-site actions.

Solution: connecting AdWords to Google Analytics

To link AdWords to your Google Analytics account, simply:

  1. Sign in to your Google Analytics account
  2. Click Admin, and select the Property you want to add AdWords
  3. Select the ‘AdWords Linking’ tab
  4. Click + New Link Group
  5. Select the AdWords account or accounts you want to link
  6. Click ‘Link accounts’

Solution: connecting Search Console to Google Analytics

To link Google Search Console to your Google Analytics account, simply for steps 1-2 as above, then:

  1. Scroll down and select ‘All Products’
  2. Click ‘Link Search Console’
  3. Once in ‘Search Console Settings’, click ‘edit’ and select the relevant Search Console Property you want
  4. Under Search Console, select the Views you want to Search Console Data to pull into
  5. Click Save
Search Console Settings

View and Property settings are just one part of the puzzle…

I hope this post has been helped you understand and feel more confident with some of common View and Property level configuration issues in Google Analytics

My next post will look in more detail at common code deployment issues and provide you with the know-how to diagnose and fix these using a number of clever auditing tools and processes.