Online Google Analytics Audit Tool

Looking for an automated tool to do your Google Analytics audits and data quality checks? Look no more!

Skip the text and go straight to the auditing tool.

Part of my job is to do quality checks of a lot of Google Analytics implementations. In doing so, I find myself manually checking for the same things over and over. So, by leveraging the Google Analytics API, I have developed an online tool to help me. And to help you!

So far, the tool runs through a dozen or so common checks. But it’s in beta, and it’s a work in progress, so I will keep adding more checks.

As of now, the tool lets you select an account, property and view to audit. It will automatically check for common errors. And will give you some pointers as to how to fix your errors (or at least what to consider).

Included Checks

Added on August 15, 2017:

  • Mismatch between configured timezone and “primary country” of traffic
  • Multiple hostnames. Did you forget to exclude development/test hostnames?
  • Self referral traffic. Are any pages missing the tracking code?
  • Exclusion of bot and spider traffic
  • Spam referral traffic. Google is getting better at blocking referral spam, but some might still slip through.
  • Crawler traffic. A kind of spam traffic that can have huge impact on direct traffic and bounce rates.
  • Basic site search settings. Like, it is working?
  • Advanced site search settings. Is this working too?
  • Ecommerce settings (standard and enhanced). Basically just if it’s enabled.
  • Duplicate ecommerce transactions. Which is a very common error.
  • Mismatches between revenue and product revenue. If you can’t trust your revenue data, you really can’t trust your data at all.
  • Presence of Personally Identifiable Information (email addresses only as of yet)

Added on August 17, 2017:

  • Goal verification. Do active goals exists, do they have goal values and are they counting?
  • Site Search Quality. Checking if keywords appear on multiple rows because of different letter casing
  • Adwords Linking. Apart from just checking if Adwords is linked to Analytics, the tool will check for common errors in each linked account (such as large clicks vs. sessions discrepancies, wrongfully linked accounts etc.)

Added on September 4, 2017:

  • Currency settings for ecommerce. Added check for current currency setting and hints + links to help article if you’re accepting multiple currencies.
  • Branded vs. Generic Paid Search. Now checking if Paid Search (as a channel group) has been split into Branded vs. Generic sessions.
  • Not set landing pages. Checks if some landing pages are (not set), which usually indicates that initial pageviews are firing incorrectly or out of order.

Added on September 8, 2017:

  • Now checking the number of configured views. Best practice is to have at least a main view, a raw view and a test view. You’ll get a warning from the tool, if you only have one or two views. And help/guide is included.
  • Adjusted check for Adwords Account Links to check for cases where you actually do have Paid Search traffic, but have “forgotten” to link your Adwords Account to Analytics

Added on September 12, 2017:

  • Added check of default channel groupings. Fails your setup if more than 5% of traffic is from (Other) and throws a warning if +0% - 5% of traffic is from (Other). This indicates incorrect or missing UTM tagging.
  • Also now checking for landing pages with very low bounce rates, which most often indicate incorrect implementations of Analytics tracking.
  • The tool now also does a quality check on the All Pages report. It examines if some pages occur on multiple rows. This usually happens if individual pages can be accessed with mixed letter casing in the URL or if a page can be accessed with and without a trailing slash.

Added on October 30, 2017:

  • Added check to see what percentage (if any) of sold products are missing a Product Category.

It’s in beta!

As mentioned, the tool is currently in beta. So if you discover errors, things you don’t understand, or have a request for a type of check, please let me know.