Don’t Trust Your Google Analytics Data Just yet, Perform a Tag Audit

Google Analytics and other equivalent web analytics solutions are valuable sources that allow online marketers, publishers, advertisers, website owners and relevant digital stakeholders to learn about the online behavior of the website audience. Easily translating data into practical insights and action items –  and be able to do so on a daily or even hourly basis – can act as a magical power-pill and a huge business strategy advantage.

To become fully data-driven professionals and be able to test, optimize and/or personalize our web properties, we need to create a solid ground of data-driven decision making environment. For that, we need to surpass a few obstacles on the way up to the land of sweet data, milk and honey. In this article, I will discuss Google Analytics in particular as a digital measurement tool that can make our day-to-day measurement tasks much easier, when implemented and used correctly.

During the last few years, the Google Analytics product team has been working hard to improve the tool, and enhance it with more features and integration abilities. In 2013 alone, they released more than 70 product updates, including cost data upload, an attribution modeling tool, dimension widening, goal templates, new admin settings and so on. The result is a very flexible and powerful tool, which holds advanced measurement capabilities for different depth-of-interactions and access points with our business, around multiple devices, online or even offline environments (with the Measurement Protocol).

On the other hand, the tool has become much more complicated to use than in the past, requiring almost “super-user”-like abilities. Making the most out of the tool has become quite of a challenge for many users. Today, some feel like the depth of knowledge that can be acquired from the tool relies heavily on implementation abilities and technical understanding of the analytics space.

A lot has been written about how to build a bulletproof Analytics implementation in order to avoid or minimize any data inaccuracies. The truth is that, even if implemented correctly, there are quite a few reasons for having inaccurate data in Google Analytics, such as sampled data, JavaScript errors, unknown “direct” traffic and so on. Even so, you can take some relatively simple actions to fix some of the common data inaccuracies by yourself.

Based on my humble experience over the years, as an avid Google Analytics user, advisor and a Google Analytics Qualified Individual, I’ve concluded a list of what I consider to be the top technical quality assurance tag audits to go through for ensuring data accuracy and better decision making.

Here’s my version of how to ensure accurate tag implementation for better insights and traffic analysis.

A Practical Guide to Google Analytics Tag Implementation Audit:

Locate missing tags

Making sure you’ve implemented the Google Analytics tracking code in all of your pages is definitely one of the most important tips you’ll get. I would also warmly recommend including the tags in any error handling pages (such as 40x or 50x errors). There are several tools that can help you with verification; my favorite one is Screaming Frog. The Frog is a spider tool, which is capable of crawling a complete root domain while searching source code for internal HTML pages for relevant custom strings, such as the Google Analytics UA number. There’s a limited free version of the tool, but, unfortunately, you’re required to pay for this specific search feature. All in all, it’s a very efficient tool for small- and medium-size websites. If you’re looking to gain more valuable information about your tracking data, there’s the Chrome Browser extension Web Analytics Solution Profiler (created by the brilliant Cardinal Path). It’s a debugging tool for data sent via tags and beacons, which is added as a new tab in the Chrome developer tools.

If you like extensions like myself, try using the GA Debug Chrome extensions. It’s a debug version of the Google Analytics JavaScript code, which prints useful information to the Chrome console tab.

All of these extensions will provide you with information from the Google Analytics tag, including the tracking account ID, hostname, current URI, page meta information, browser information, executed events and so on. Of course, you can also use the Google Analytics Real-time reports to see whether the pages you’re browsing actually appear in the report. For large scale audits, use tools like GA Checker.

Identify wrongly implemented tags and/or inconsistency in tag versions

We should pay attention to two things:

First, there are different versions of the Google Analytics tracking script. There’s the traditional synchronous script versus the asynchronous script (both based on ga.js), there’s the Universal Analytics script (analytics.js), a Google Tag Manager based script (gtm.js) and, if you haven’t updated your Analytics in a long, long time (shame on you…), there’s also an Urchin tracking script (urchin.js). Make sure all of your web pages use the same version of GA script.

Second, if you’re using the traditional script (ga.js), you may need to add a line of code to your tracking script for tracking and collecting data from multiple domains in a single view. If you do need to do that, make sure you’re modifying the script in all of your pages.

Eliminate multiple tags on web pages

I’ve seen too many cases where there were forgotten pages with multiple tags running in them. It’s very straightforward advice: use only one tag in a page, unless of course you’re trying to execute multiple properties in a single page and you know what you’re doing. You can use the free Chrome extension: Tag Assistant (by Google) to investigate pages you’re visiting for any tracking errors or suggestions.

Investigate self-referral traffic

Another great methodology for locating untagged pages is by investigating self-referral traffic in the Google Analytics acquisition reports. Self-referrals are traffic to your site from your own website. There are many common reasons for having self-referrals in Google Analytics. There are always legitimate reasons for having self-referral traffic, but, in many cases, it happens due to implementation errors. For example:

• The referring page is missing the tracking code, but the tag exists in the landing page.
• Tracking code implementation inconsistencies can cause multiple cookie sets, or even override of cookies.
• Meta-refresh and JavaScript-based redirects, especially when loaded before the actual execution of the Google Analytics tag.

So, in order to identify those pages, go to the Acquisition menu, choose All Referrals and look (or filter) your own root domain. Now, click on your domain and start investigating all of the referral paths. Some of them may be missing the Google Analytics tag.

In Summary

Tag auditing is the first step (of many others) toward ensuring data accuracy. I would recommend conducting an analytics tag audit at a monthly basis. There are many tag auditing tools and methodologies available, but I hope that the ones suggested here will help you audit your tags and make sure that data is recorded properly. Happy analyzing.

Read part 2: Don’t Trust Your Google Analytics Data Just yet (part 2): Get Rid of Noise With These Fundamental 9 Filters