Our developer Tom Maddocks, who you may recall from his rather dubious “a day in the life" feature, shares his top tips on making the most of your Google Analytics. In this feature, we promise he’s telling the truth.
Over to Tom...
Recently I’ve been working more closely with our ecommerce clients to make sure their Analytics are performing the way they should. As it was Easter recently and we’re all in the mood for new starts, I thought I'd share some potential problem areas that I’ve spotted and give some advice on how to resolve them.
It’s important to know where your traffic is coming from but are self referrals skewing your data? A self referral is where you see your own domain in your referral reports. To check this head over to your Analytics reporting and select the following from the navigation: Acquisition > All Traffic > Referrals
Self referrals can occur for a number of reasons, the main ones being:
- Tracking code missing on certain pages
- Incorrectly configured cross domain tracking
Missing tracking code - if you’re missing tracking code on your pages then this will cause a session to end, so when visitors navigate to the next page with the correct tracking code, a new session will begin. These should be easy to identify in your landing page report if you filter by full referrer. See the tools I use at the end of the article to highlight if a page is missing tracking code.
Incorrectly configured cross domain tracking - your setup may be wrong if the landing page and full referrer are on separate domains (shop.com and checkout.shop.com). To narrow these down head to the landing page report: Behavior > Site Content > Landing Page Report
From here apply a secondary dimension of Full Referrer. This way you can see when the traffic arrived and where it came from.
Finally, apply a segment for Referral Traffic.
If you’re still seeing your own domain(s) in here then you may need to configure your cross domain tracking correctly. If this is the case, you can read more here:
Another issue ecommerce clients face are referrals coming from domains that shouldn’t be recognised as referrals. The classic case in this instance is third party payment providers, e.g. PayPal.
If you haven't set up your referral exclusions, then each time a customer pays with a third party provider and then returns to the checkout confirmation, a referral will show from the payment provider. Obviously this can wreak havoc on your data and it can mean you lose the true referrer of the traffic, e.g. organic, PPC etc.
So how do you exclude this kind of traffic? Simple. In the top navigation of your Analytics account head to: Admin > Tracking Info > Referral Exclusion List
In here, just click Add Referral Exclusion and enter the domain you want to exclude.
Funnel tracking is probably one of the most important aspects of good ecommerce tracking and is vital in understanding where users drop off throughout any step by step process, with the checkout process being one of many examples. As an added bonus (especially for us developers) it can help highlight potential bugs across different browsers.
Imagine we take a look at the drop-offs throughout the funnel and notice an unusually high number of customers using the same browser version. This would certainly warrant further investigation. Are these users seeing something different to users on more modern browsers? Can’t they interact with the page in the same way? Without Analytics it’d be harder to identify problems like this.
A full guide is out of scope for this brief article but here's one I found earlier - Eric's produced a great guide, check it out: Conversion funnel survival guide
So what tools can you use to help test your Analytics and highlight potential pitfalls? A couple that I use are:
Google Tag Assistant is especially useful as it helps you troubleshoot various Google tag installations, including Google Tag Manager, Analytics and more. What’s particularly good about this tool is that it suggests improvements that can and should be made to your tag implementations. It’s also a bit more user friendly for the non-tech users.