Are you sure your Google Analytics (GA) setup is tracking all the right data about your e-commerce site? If not, the analytics reports that you are looking at are maybe inaccurate.
Only an accurate analytics data can help you make informed business decisions, bring down costs, and offer valuable insights about your e-commerce products and services.
But how do you ensure that Google Analytics tracking data precisely? By carrying out a Google Analytics audit. An audit will make sure that the tracked GA data pertaining to your e-commerce site is trustworthy, accurate, and actionable.
So, here are a few useful tips to help you conduct a GA audit.
When should you conduct a Google Analytics audit?
Some of the usual scenarios where you need to ensure the quality of your analytics data are,
- After the initial setup or website launch
- Website migration
- After an overhaul of the site’s structure, CMS (Content Management System)
- After amendments to the business goals and/or business logic
It is usually recommended to perform multiple audits every year on your site, even if just one of the above conditions is true for your website. Also, take care that effective communication happens between different departments of the organization to reduce the odds of poor communication resulting in inaccurate analytics data.
The best tools to perform a Google Analytics audit
Here are some common tools,
Chrome Tag Assistant – This is a Chrome Extension that offers a brief overview of the installed tags on any web page of a site and also if they are working properly.
Chrome Google Analytics Debugger – This is also a Chrome Extension. It loads the Google Analytics code of a site that you browse in the debug mode in a console. This console displays error messages and warnings related to a faulty Google Analytics setup. It can be a tad overwhelming to look at these messages and figure out their meaning if you have no programming background.
Automated Tool – If you need to perform several audits every month on multiple websites, it is recommended to have an automated audit tool to make the task easier.
The areas to examine in a Google Analytics audit
Here are the most critical areas to inspect when carrying out a GA audit.
Google Analytics Code
Whether the GA code has been set up manually or with a tag management tool, make sure that it is present on all the web pages of your e-commerce website. GA Checker is a good tool that I would recommend for this task.
Set the GA implementation in such a way that it is not counting the traffic from your own domain (self-referrals) as external referral traffic. To set this up, add your domains to the Referral Exclusion list. You can go to this Google web page to know more about Referral Exclusion. You can go to the Source section to check if your own domain is showing up on the referral report.
To ensure the integrity of your GA code, check whether all the relevant events on your site are being captured successfully. The Google Analytics code must track a variety of pre-conversion/purchase events also such as adding an item to wish-list, completing a survey, signing up for the newsletter and so on. Recording these pre-purchase events are just as important as tracking the purchases themselves.
You can view the report on the tracked events in GA by logging into your GA account –> Select the desired Account/Property/View from the Admin section.
Then navigate to Reports –> Behavior –> Events to generate different reports related to the Events.
Version of GA Code
Check if different sections of the whole website are having the same version of the Google Analytics code. It shouldn’t happen that few sections are implemented with one version of the GA code and the others with another version. Such a scenario can arise if different sections of the website are managed by different divisions of an organization. This version disparity can result in faulty analytics tracking, broken sessions and an exorbitant amount of self-referrals showing up in the reports.
Check if you have added two Google Analytics scripts for any given web page. If that is the case, it will count every metric twice (page views, visits etc.) on your website and will also lower the bounce rate significantly (to less than 3 percent or so). This unnatural drop in bounce rate is a tell-tale indication of double-tracking. If you suspect double-tracking, use a crawler to identify the issue and remove it.
You shouldn’t be tracking the traffic from domains other than what you have registered as the ‘Property’ in your GA account. If you are seeing other websites (or hostnames) in the report, you shouldn’t be tracking their traffic and must get rid of them by removing the GA code from them or blocking the traffic from those sites from ‘View’ option, which you can access through the Admin button.
You can get to this report by navigating to ‘Acquisition’ –> ‘Source/Medium’ –> ‘Other’ (in the report screen under the graph). Please check the below screenshot.
Make sure you remove the GA Code from all the development sites or sub-domains to exclude their test traffic from your live site’s metrics. Even though the percentage of test traffic from your development domains is minimal, it can lead to inaccurate metrics tracking of your actual site. And while setting up filters, use them with caution, as some filters can block GA from tracking critical analytics data about your domain.
Site Search Feature
If your site offers an Internal Search feature, check that you have turned the ‘Site Search Tracking’ feature ON under the View Settings (in Admin –> View) in your Google Analytics account.
The benefits of providing internal site search feature are numerous from an e-commerce analytics perspective. Tracking what users are searching for within your helps you decide what new products or related products to introduce in your e-commerce site, tweak how products should be arranged or (product visibility) on the site, measure product demand and offers a variety of other useful and actionable analytics data.
You must verify that traffic from the users working on your e-commerce site is not being tracked by Google Analytics. This traffic can be excluded from being tracked by GA by blocking an IP or a range of IPs using ‘Filters’.
You can access these filters by navigating to Admin –> View –> View Settings –> Filters –> Add Filters. Please refer to the following screenshot.
The Filter Type must be ‘Predefined’. Select ‘Exclude’ option from the ‘Select filter type’ drop-down box, then select ‘Traffic from IP address’ option from the ‘Select source or destination’ drop-down and then select the appropriate expression (‘that are equal to’ or ‘that begin with’) from the ‘Select expression’ drop-down box.
More information on IP filters can be found on this Google Support page. You can either block a single IP or a range of IPs using the filter option.
Confirm that your ad campaigns (that you can edit) or referring sources are appended with UTM tracking tags. This lets you understand which ad campaigns (Google Adwords, Email Marketing etc) or referring sites are performing well, and which are NOT. You can see your sources and referrals by navigating to Reports à Acquisition à All Traffic à Source/Medium and/or Referrals.
UTM tracking-variables within a link allows you to track superficial info like visitor count from a social media account to finer details like revenue generated from your tweets of your campaigns/sources. But while adding tags to your links take care that you don’t use different tags to track the same link, else it will make deciphering the reports cumbersome. If you are still wondering what UTM tags are here is a nice guide to get you started.
Goals and Goal Funnels
There must be at least one macro goal and one or more micro goals set up for your e-commerce sites. A macro goal is an ultimate goal that you want to accomplish on your site, i.e. a successful completion of a purchase or a sign up to an email list.
Micro goals can be any of the secondary actions that the users take on your site such as interacting with the Live Chat feature, adding a product to the cart, adding a product to the wish list, searching for a product in the internal search and so on that leads up to the ultimate macro goal.
And if it is relevant, you must also set up goal funnels. A goal funnel will have the steps or stages that your customer must go through to accomplish a goal that you have defined. For instance, if you have set successful purchase of an item as a goal, its goal funnel can have the following steps.
Step 1: Product search,
Step 2: Adding it to the cart and
Step 3 (final step): The purchase of an item.
This will let you know which steps need more optimization and which are performing at its best. Goal funnel also consolidates the data about multiple actions/areas in one place under one report, which is very convenient for an e-commerce site owner.
(Image courtesy Megalytic.com)
Visualizing a goal funnel in Google Analytics will be especially useful if your e-commerce check-out process has multiple pages and upsell products after every purchase.
Payment Gateway Referrals
Most e-commerce sites are set up in such a manner that when a customer completes a payment on a payment gateway (for e.g. PayPal) after a sale, the customer is sent back to the vendor site. When this happens, Google Analytics starts a new session with the payment gateway site as the referral and overrides the previous traffic source (referral site) that brought in the customer to your e-commerce site. This prevents you from seeing the actual referral sources sending you successful sales.
Hence, it is critical to check if payment gateways are being inadvertently captured by the Google Analytics implementation as a traffic source for sales. If that’s the case, you must add it to the Referral Exclusion List under the Property pane.
You can navigate to the Referral Exclusion List page by going to Admin à Property: Tracking Info –> Referral Exclusion List. Please refer to the below screenshot.
Backend and GA data
The most critical and the rudimentary check that you must perform as part of a Google Analytics audit is comparing the actual backend data with the GA data. Compare the orders, the total number of sales, and total revenue on GA with the backend data, and validate if they really match or there are discrepancies. The comparison shouldn’t be limited to just the successful orders but all the orders that you received. If the difference is negligible, it won’t make a huge impact on your analytics data and can be ignored. Small discrepancies do happen at times and are inevitable. It must also be tested that the time-zones of both the backend and GA data are the same. To troubleshoot the discrepancies, it is necessary to carefully compare each transaction in the backend with the GA data.
By the way, if your Google Analytics auditing requirements are much more complex, it is highly recommended to Turn on the Enhanced E-commerce Reporting feature in Google Analytics. By setting up this feature, you can track a wide variety of insightful e-commerce metrics pertaining to your sites such as clicks, product impressions, the performance of your checkout funnel and internal promotions and you can even keep track of your refunds.
You can access this feature in Admin à View –>E-commerce Settings
But after you set up the Enhanced E-commerce feature your GA auditing also becomes much more intricate as you need to check a variety of parameters across the whole site in the audit process.
Wrapping it up
You must understand that a detailed Google Analytics audit must be investigative in nature and shouldn’t be limited to only the crucial reports or setups of the GA setup as discussed in this article. Also, different sites can have varying GA audit requirements depending on their metrics analytics needs.
Basic metrics tracking will only require a straight-forward audit, but if it goes into finer details and involves interaction with other tools, the audit must be well-laid out and detailed, for e.g. In a scenario where the data generated by the triggers and tags created in Google Tag Manager are fed into GA.
Hence, to make the GA audit comprehensive and thorough, you must make a checklist of setups/reports that you need to validate to make the process easier and efficient.