Google Analytics Audit Guide

A bad Google Analytics (GA) audit is as useful as an inflatable dartboard.

But a robust Google Analytics setup helps you to make smarter, more profitable, ecommerce decisions. Google Analytics Audit Guide

This article will cover:

a) Limitations of automated GA checker tools

b) How to check if GA is working (a roundup of common problems, testing tools & techniques to debug).

c) How to do proper GA implementation

d) What to look for if outsourcing this work to a GA consultant

A Google Analytics audit is a way of ensuring that your website is:

a) Tracking accurately

b) Tracking comprehensively (all the data that’s important)

In practice, it involves editing your website code to send the right data (at the right time) to the Google Analytics servers and reports; it also involves configuring the admin setup inside of the GA interface.

Your web analytics audit should start with the end in mind. Determine where you want to get to (output). This will make it much easier to determine the tracking requirements (inputs).

You can determine the outputs with a measurement model. A measurement model is a one-pager that outlines the business, marketing and website KPI’s that you want to track. It takes into account the dimensions and metrics, paths to conversion (funnels) and traffic sources that matter most to you.

For example, a digital measurement model for a retail website could look like this:

Digital Measurement Model For A Retail Site
(furniture website example)

GOAL Improve Visitor-To-Customer Ratio Increase Free Fabric Samples Increase Newsletter Signups
MACRO KPI Revenue Per Visitor conversion rate ‘Free Fabric’ Signup conversion rate Email Opt-Ins
MICRO KPI’S Product List Page conversion rate Internal Traffic To Free Sample Page Email Opt-In Rate
Add-To-Cart Rate External Traffic To Free Sample Page
Checkout conversion rate
Average Order Value
Returns Rate
Lifetime Customer Value
SEGMENTS Organic Search
Facebook Ads
Paid Search
Organic Search
Facebook Ads
Paid Search
Organic Search
Facebook Ads
Paid Search
Dimension 1 Subscribers vs Non-Subscribers
Dimension 2 Customers vs Non-Customers
Dimension 3 Samplers vs Non-Samplers

Table 1: creating a measurement model helps to determine the metrics to be tracked on your ecommerce website.

The above example is a 10,000 foot view. You can dig deeper with more granular metrics that support the above metrics.

For example, your add-to-cart rate might be improved by increasing video watch rates on your product page, so you’d want a metric that tracks video plays, and a dimension that tracks ‘video players’ vs ‘non-video players’.

Or, in order to track your overall checkout conversion rate, you’d want to track the individual steps within the checkout journey; each step could be a standalone conversion metric.

There is no cookie cutter template for this, but you can copy the above table to get started.

To create your measurement model, work backwards from your business KPI’s and translate these KPI’s into ecommerce and website metrics.

GA Checker Tools

GA Checker Tools have a lot of limitations because there are so many Google Analytics ‘gotchas’, and because the microeconomics of your business are unique.

For these reasons, verifying your Google Analytics setup is a very hard task to automate.

Several GA Checker Tools have been developed (overview below), but bear in mind that none of them comprehensively support your measurement model on their own.

GA Tag Check

GA Tag Check tools are automation tools that scrape the Google Analytics code on your website and check for errors.

GA Checker

Crawing a website with

Running such tools may answer questions like:

Is the Google Analytics code installed on every page?

Are hits being correctly sent from the website to the Google Analytics servers?

Is the GA script located in the write location in the page source code?

Is the code syntax valid?

Is there any duplicate code?

How fast is the GA code loading on the web pages?

Are there any generic dataLayer syntax errors?

In addition, some of these tools offer generic recommendations about how to fix these issues.

Popular GA Checker tools in this vain are:

A cloud-based web app that crawls your entire site for missing Google Analytics code (and other Google tools). All you have to do is enter your domain and submit. The tool simply checks for the presence of your Google Universal Analytics (analytics.js) code. If your Google Analytics script is installed via Google Tag Manager (GTM), you would need to check for the presence of gtm.js on your website.

In the case of Google Analytics, verifies the presence of a UA number and track pageview call (where applicable). In the case of Google Tag Manager, it checks for the presence of the javascript file.


– The tool isn’t really maintained. It’s buggy. It runs on a non-secure website.

– It doesn’t guarantee your code is error-free, just whether or not the code is there.

– It doesn’t tell you about which GA or GTM properties are tracked.

– It doesn’t tell you anything about the data layer (more about the data layer later).

– It doesn’t know anything about your custom measurement model.

Powerful website crawler that runs as a desktop application. It crawls your website pages. Originally an SEO tool built for the purpose of technical onsite SEO audits, it can also be configured to check for whatever script you like (like Google Analytics).

Screaming Frog

Screaming Frog’s desktop application

Here is a simple how-to guide to check for your Google Analytics code (the same process can be used for Google Tag Manager).


– Easy to search for specific GA and GTM code snippets.

– More powerful and more reliable than

– It doesn’t guarantee your code is error-free, just whether or not the code is there.

– It doesn’t tell you anything about the data layer.

– It doesn’t know anything about your custom measurement model.

This is a Chrome browser extension that helps you to troubleshoot installation of various Google tags including Google Analytics and Google Tag Manager.

Install the extension, navigate to one of your web pages and Tag Assistant will tell you which tags are present, report any errors and suggest improvements that can be made to your implementation.


– You can only spot-check on a page by page basis; it doesn’t crawl your website like

– It reads the tracking script, so you can see exactly which UA properties or GTM accounts are installed.

– It reads the on-load dataLayer but not data layer variables that are ‘pushed’ after page load.

– It doesn’t know anything about your custom measurement model.

In summary, automated GA checker tools are a helpful starting point for Google Analytics audits but lack crucial information to really troubleshoot your setup.

You need to go deeper in order to understand information like:

– Whether you are tracking all the funnels, buttons, video plays, opt-ins, downloads, etc. that matter to your business

– Whether enhanced ecommerce reporting is set up, and tracking correctly

– Whether the dataLayer is sending accurate metadata to Google Analytics

– In short, is your measurement model being tracked?

WASP is another tool that speeds up the audit process. I don’t really classify it as an automation tool because it still involves manual review:


WASP’s user interface is an additional tab inside Chrome Dev Tools

WASP is a quality assurance tool for web analytics tags. It runs on your Chrome browser as an extension, scanning your website for all marketing tags, including Google Analytics.

WASP is more comprehensive than any of the aforementioned GA Tag Check tools because:

a) it has a smarter algorithm which auto-detects every type of tag more reliably and with more granularity than standard automation tools.

b) it has a smart visualisation to help you read all the data.

WASP is a powerful way of getting in-context information about the tags present on a page as you browse.

Here’s a how-to video:

(This video is for WASP.inspector rather than WASP.crawler. WASP.inspector is the Google Chrome browser extension. WASP.crawler does the same thing but isn’t limited to a page level. Rather, it crawls your website and returns the data into a neat format that you can read later.)

GA Admin Tool Checkers

GA Admin Tool Checkers are an entirely separate but complimentary class of tool. Rather than scrape your website code, they scan your GA admin setup (i.e. what you see in your GA account interface) for common problems.

Such tools require authentication so that they can access your GA account.

Once again, the nature of the automation is superficial. They’ll tell you information like:

Whether you’re capturing too many events in your GA account (which can slow down website performance).

Whether you’re capturing any custom dimensions or custom metrics.

Whether you’ve enabled certain features like content grouping, enhanced ecommerce or demographic reports.

Whether your GA views are set to filter out irrelevant traffic

Basic GA housekeeping that is generic to ALL websites but not specific to YOUR website.

GA Account Check

Some tools scan your GA admin setup and reports for common ‘housekeeping’ errors.

Here are two tool examples:

1. (free)

2. Brain clifton’s Verified Data: (paid)


– The landscape is changing all the time and, personally, I don’t trust these tools to be updated regularly.

– For more comprehensive results, consider running multiple GA Admin Tools to gather as much feedback as possible.

– This type of audit needs to be paired with a GA Tag Check in order to understand how the data is being sent to Google Analytics in the first place.

– The tools don’t know anything about your custom measurement model.

– Checking your GA admin setup manually is not all that time consuming in any case. See the below section called “Google Analytics Audit Checklist”.

Data Layer Inspection Tools

Quick Recap On Data Layers

“Tag management systems have become a prerequisite for effectively tracking your website – BUT they have one massive flaw. Their true power can only be realised when a complete and valid data layer exists on the website.”

A dataLayer is a javascript object that is used to pass information from your website to your tag management solution (like Google Tag Manager).

The concept of a data layer is integral to a robust Google Analytics setup.

In a web analytics context, the data layer is the behind-the-scenes structure on your website from which data is sent to Google Analytics.

It has to be built by your web developers as it has to pull data from your website server. Only web developers who can manipulate the server-side code are able to implement the data layer for you.

If you’re using a tag management solution like Google Tag Manager (and you should), and you’re using a data layer (and you should!), then you need to audit your data layer.

As an example, a data layer on your order confirmation thank you page might contain the following four variables:

Product: Nike Air Jordans

Size: 10

Price: $44.99

Colour: Blue

Data Layer Inspection Tools are Chrome browser extensions that allow you to inspect the dataLayer in real time so you can see what variables are being set and on what events.

With that said, Data Layer Inspection Tools are optional, since it’s low-effort to ‘look under the hood’ at your website code to review the data layer for yourself. You can do this by opening Chrome Dev Tools or Firefox Developer Tools.

Or by using the preview/debug mode in GTM:

GTM Debug Mode

Inspecting the Data Layer using Google Tag Manager

The advantage of Data Layer Inspection Tools is that they offer a better user experience than anything else, since they have the sole purpose of reading the data layer.

These days, there are a lot of data layer inspectors out there, including:

Data Layer Checker – by

Dataslayer – by Sean Adams

Data Layer Inspector+ – by Analytics Pros

Data Layer Checker


By now, it should be obvious that no single tool is a silver bullet for your Google Analytics audit.

GA Tag Checkers, GA Admin Checkers, and Data Layer Inspectors need to be used in combination not independently.

Using some of these tools in combination can support and accelerate the auditing process.

The advantages of such tools is that they can spot code errors more quickly and accurately than a human. They are a good starting point for a Google Analytics audit.

The crux is that no tool intrinsically knows what data you should be sending to Google Analytics. They can tell you what’s there, but it’s up the human auditor to make a judgement about the nuances of the data that you are capturing. This MUST relate back to your measurement model.

Google Analytics Audit Checklist

Over the years, I have built and regularly update my own Google Analytics Audit Checklist. I use this for all my GA Health Check projects for ecommerce clients, and update it whenever things change.

This checklist shows you what to check in Google Analytics (email address required). Familiarity with Google Analytics itself is extremely beneficial, but if you’re not an expert then I’ve hyperlinked to how-to guides where applicable.

With this checklist, you can go through your Account, Property, and View settings and make any necessary changes.


Make friends with the Google Analytics Admin Interface

The checklist is also a good starting point for implementation (see below section called “Google Analytics Implementation”).

Again, you can access the Google Analytics audit template by clicking on the button below:

(your Google Analytics implementation checklist)

How To Check if Google Analytics Is Working

To check if Google Analytics is REALLY working you need to know that:

1. It supports your measurement model

2. That are no tracking errors

3. The GA interface is configured in a way that supports 1 and 2 above

Google Analytics out of the box seems relatively straightforward. Place a piece of code on all your web pages and watch the data roll in.

The reality is that you need to configure it to tracks events, goals and ecommerce data. You also need to make adjustments in order NOT to track non-relevant data like spam or internal sessions.

When conducting my own Google Analytics audits, I start by using some of the aforementioned automation tools. After the automation, I go down the manual route.

A manual audit is the only way to ensure that your measurement model is comprehensively supported.

Robots Lack Acumen

A manual audit involves:

1. creating your measurement model;

2. reviewing your website tracking code;

3. reviewing your GA Admin Account.

You can use whatever tools you like to achieve this (I’ve recommended some best-in-class in this article), but here’s what I do for all my Google Analytics Health Checks.

A manual audit involves:

1. Create a Measurement Model

2. Use Screaming Frog to determine if all the GA and/or GTM code is there

3. Run Google Tag Assistant for page by page analysis to identify code errors on the most critical pages e.g. ‘path to purchase’ pages

4. Inspect the dataLayer on the primary pages (I prefer Data Layer Checker for this)

5. Review the GA admin setup using my GA housekeeping checklist

6. Use WASP (Web Analytics Solutions Provider) to conduct a thorough tag check 

What you’ll notice is that this is fairly manual work! Tools help, but you just can’t automate this process. At least not yet. Because the tools don’t understand your custom tracking requirements.

The ONLY way to comprehensively determine whether your Google Analytics is working is by looking at the code and the GA admin; then cross-referencing this with your measurement model.

The output of the above process is an understanding of what’s missing from a data perspective. You can then move to the ‘implementation’ phase to make your fixes (see next section!).

Google Analytics Implementation

Google Analytics implementation is the work that you do AFTER you have audited the current GA tracking setup.

Implementation work involves configuring the GA admin interface, and configuring the website code itself to control what data is sent to Google Analytics.

To configure the GA interface, you can download my housekeeping checklist here. Honestly, with enough GA knowledge, changing the admin interface is straightforward. Even if you’re not a GA wizard, if you have time, there is a wealth of online information about how to update Google Analytics settings.

The second part of the implementation work is MUCH harder: changing the website code to support your tracking requirements.

Data Layer Instructions Excerpt

Excerpt of data layer instructions for a previous client

As part of the audit process, you’ll probably identify several code updates that need to be made. Most of these changes will be supported by configuring the data layer. All of them should be supported by a tag management solution like Google Tag Manager.

To make implementation changes on the website, you need to write javascript and data layer instructions that your developers can use. Remember, the data layer is integral to a robust Google Analytics setup. It’s the behind-the-scenes structure on your website from which data is sent to Google Analytics.

Data Layer

Hello, data layer!

Since the data layer is a javascript object that is used to pass information from your website to your tag management solution (like Google Tag Manager), it has to be built at the same time that the page HTML is being built – on the server.

Someone with javascript expertise (and knowledge of how to collect data using analytics.js) will write these javascript instructions.

Then your web developers will manipulate the server-side code to implement the data layer for you.

Essentially, the dataLayer is populated with variables, from your internal website server. For example, say that someone buys a pair of shoes and has just made the following purchase:

Product: Nike Air Jordans

Size: 10

Price: $44.99

Colour: Blue

That’s four variables. On the purchase order confirmation page, the dataLayer would be populated with those variables. In turn, this data would be picked up by Google Tag Manager and sent to Google Analytics.

That’s just one example. Examples of other variables include:

– Whether the visitors has subscribed to your newsletter (subscribers vs non-subscribers)

– Whether the visitor is an existing customer (customers vs non-customers)

– How much the visitor has spent before (lifetime customer value)

– And much, MUCH more! Seriously, MUCH more!

There’s no way for Google Analytics to know this information unless you tell it. It has to be built into the data layer.

What do data layer javascript implementation instructions look like? Here’s an example for one of my previous clients. Bear in mind that your own dataLayers will be unique to your website. This is just to give you a feel for it.

You can also take a look at the official Google documentation – but to be honest it’s notoriously convoluted and unhelpful. You can learn quicker by reverse-engineering an existing setup, such as on the Google Merchandise Store.

Datalayer Checker

Inspecting the datalayer on well-configured websites like the Google Merchandise Store is a shortcut to learning.

Bear in mind that this article is not a javascript or a data layer tutorial. My objective is to give you the framework so you know what are the necessary steps to conduct an effective Google Analytics audit.

Google Analytics Consultant

Writing these javascript requirements is one of the main reasons that Google Analytics Consultants are often required. You need someone to write the dataLayer. Then, once it’s been implemented on the website by your developers, you need someone to check that implementation.

Most Google Analytics audit services will include this as part of the service. Helpfully, Google has a list of certified companies who know the ins and outs of Google Analytics (the Partner Gallery). These companies offer services ranging from consulting to implementation to tracking.

Unfortunately, as a solo consultant, I cannot feature the Partner Gallery (!) – Google requires a minimum headcount of five. But you can apply for a Google Analytics Health Check with me here.

Google Analytics Expert

What should you be looking for from a Google Analytics expert?

1. Someone who starts with the end in mind. They understand your business/ecommerce requirements, and don’t jump straight into the tool. Google Analytics is blunt instrument that requires custom configuration.

2. Google Analytics Qualification – at a minimum, your expert needs this. But to be honest, it’s not all that hard to pass the 80% pass mark. What you really should be looking for is…

3.experience. Have they done this before? How many times? In your industry?  Track record of client satisfaction

4. Methodology and templates. Clear documentation on planning processes? Frameworks and templates used to deliver complex solutions?

Google Analytics mastery requires both subject matter expertise about the tool itself AND strategic expertise to apply that knowledge to your business. This is what a true Google Analytics specialist does, and what distinguishes her from generic web analytics implementation.

“Let’s do a Google Analytics audit, it’ll be fun!” said no one, ever.

The concept of an ‘audit’ conjures up thoughts of paperwork, bureaucracy and paper pushing. But facilitating accurate website and marketing measurement helps you to:

  • enable KPI tracking that aligns your teams to improve productivity;
  • allocate your marketing budget more efficiently;
  • make smarter ecommerce decisions.

Now, that IS fun.

Oh hi, I’m Alan Chapman and you can get my free weekly email full of crafty marketing tips for ecommerce businesses here. It’s called Selling On Websites (see below).