Firstly, thank you!

I want to personally thank you for coming out and joining the other awesome speakers and myself.

Secondly, I’d like to say thanks for taking the time to check out what I couldn’t cram in to my slot.

Join me on the PPC vs SEO SEMRush Webinar – Feb 27th @ 5PM

Join me, Mike and the SEMRush team in a battle of the disciplines (this time SEO vs PPC) we’ll be going in at each other but also dropping knowledge bombs on how teams in PPC and SEO can work better together.
Click here to learn more

Anyway, I promised you guys a dev-free report. I’ll reiterate again that it’s worked for the most of the websites that I’ve implemented it on, but not all.

For some, you will need some dev help, but the good news there is, the work needed to get it working is often very minimal.

Anyway, without further ado:

The 404 report prioritised by customer impact

So for most websites I’ve had my hands on, when a 404 error is triggered, some if not all of the page’s title is consistent/static. Out of the box, Google analytics captures the page title. Result!

This means you can use this 404 page title to identify specific 404 URLs that your visitors are bumping in to and capture other key metrics alongside, thus allowing you to prioritise by user severity and impact.

To build it, you’re going to need to build a new custom report

The good news is, it’s easy! To do that head to the Customization tab > Custom Reports, then click ‘New Custom Report’.

Next, you’ll need to specify your metrics and dimensions

This is where it’s mainly up to you to decide the data you’d like to see, but as a minimum i’d recommend pageviews, sessions and users for metrics and page from dimensions.

Note: The report is by default sorted by first metric you specify in your report. Put pageviews or the metric that’s most important to you in this spot

However to take things a step further, i’d also add in landing page and exit page dimensions too. Then when a page matches across all three of those dimensions, you can be certain that a visitor entered via a 404 and left on the same page.

A telltale sign of a potential customer getting burnt by a 404 and thinking ‘sod it’.

Levelling up again, you can add in source and/or medium columns, but one of my favourites is absolutely full referrer. This gives you precisely where referring traffic that generated the 404 has come from.

Add in a page title filter

This filter ensures that the only URLs the custom report will capture will be 404 errors To do that, you’ll need to add a filter on to the page title dimension specifically.

Here i’d recommend using regex. I’m no Regex wizard, so if I can do it, anyone can. Ultimately you want to match a string, in my case, it’s matching ‘404’. So the regex I’ve used is:

404.*

This ultimately means that if the page title starts with ‘404’ (as it does in my example) it’ll be captured in this report.

If you wanted to be a little less prescriptive and filter based on 404 appearing anywhere in the page title, you could use:

.*404.*

If you wanted to specify that the title must end in a 404, then flip the first statement around to:

.*404

Again, I’m no regex expert, so those that are might come up with a better way of achieving the same result, but this works nicely for me.

Tip: If you’d like to narrow down the number of reported rows (ideal for working on large sites, you could add in a second filter to only show you 404 errors with pageviews greater than or equal to x for a specific URL.

Finally, choose your report template and name your report

For this, you can choose to have either a flat table or a Explorer view. The explorer view is certainly more familiar and is often the template used for a lot of the built in GA reports.

For when you need to review multiple columns of dimensions and metrics for a single row of data, the flat table layout almost always works better. As that’s likely to be the case here, I’d recommend flat table, but see what works best for you.

One downside of the flat table is that there’s no line graph to stare at and the flat table view can make the report look a little more cluttered as there’s potentially more data on the screen at any one time.

However, if it’s substance over style you’re looking for, flat table is likely to be what’ll work best for you.

Once you’ve finished specifying the data and your filters, you should have a report settings screen that looks a little something like this:

That’s it! Hit save and browse through your data

I find this report hugely helpful as it’s not only flagging up the real-world 404s for a given date range, it’s able to tell you precisely how many visitors/customers are impacted by specific 404s.

The greater the number of people that are impacted by them, the higher a priority the fix. Yes, it’s less to do with SEO, granted, but it’s one step closer to creating a solid and friction-free experience for your customers. Which is arguably what it’s all about.

Final tip: If you hit the share button at the top right of your custom report, you can schedule the report to send to you on a schedule that suits so you don’t miss out on any juicy 404s that may need urgent attention!

Now for the other stuff I didn’t get time to talk about..

Ok, so I shared my favourite tips to make GA sexy when it comes to eCommerce, but there are other ways to kill it with GA in eCommerce too.

Below are some of my other favourites and go-to’s when I’m lucky enough to get hands-on with a clients GA account.

Firing events based on where users click

This one can be super insightful and helps you to understand precisely what it is your audience and customers are clicking the most often on search results, category or listing pages.

This data isn’t something that’s referenced or reviewed very frequently, admittedly.

However the data can reveal a lot about user behaviours when it is accessed and is especially handy whenever you’re considering web design revisions or updates.

Understand where most of your users click on category/listing pages.

By firing an event based on the element clicked, it allows you to understand at scale just how your users are engaging with CTAs, hyperlinked images etc.

You can understand whether visitors prefer to access your products by clicking on an image, the product title or even the ‘view product’ call to action.

You can then understand just what your user base’s preferences are and optimise accordingly.

If your users prefer to click product images to access products over any other way, maybe there’s a case to make the image thumbnail larger or invest in better product photography.

You could also work to the contrary; identifying what visitors and users don’t care about so that those elements can be optimised, made smaller to make room for more frequently used elements or just scrapped altogether.

One thing to bear in mind; just because an element is clicked a lot doesn’t mean it’s optimised for your audience and customers.

For example, just because visitors click the images the most, it doesn’t mean that clicking the image takes them to their desirable destination. Rage clicks, perhaps?

As a side note, It’s also astounding to see just how many websites (not just eCommerce websites) that don’t hyperlink their images on listing pages

Going one step further, you can go far beyond tracking click behaviour on listing pages. What about the basket and checkout? Suggested / predictive search clicks? Registration forms?

“We’re getting rid of the view product button in the new design” Now you can prove whether that’s a good move or not

You can attach these specific CTA click actions to conversion data, doing so will tell you two things:

  1. How well does the ‘view product’ CTA assist in conversions and sales. If it’s high, you might want to delve deeper before making such a change
  2. How often is the ‘view item’ CTA clicked in total? Rarely used? Removing it could clear up the design.

    3. Got an international eCommerce presence? How does click behaviour change per country?

    You can then understand whether you can offer a global UX experience to all customers moving forward or whether there are any exceptions to the updates your team is considering.

    Tracking rage clicks

    Although not always conclusive, this can be really helpful to bring to the surface the areas of your website that are frankly pissing people off.

    You can find awesome wins with something like this. Have you a worryingly high number of users punching the ‘complete purchase’ button in quick succession?

    How about identifying users that are attempting to destroy the ‘Add to Basket’ CTA?

    Sure, to get this one working will require some development input, but you’re ultimately wanting to fire an event when a single element on a web page is clicked 3x or more within around 1s.

    You want the event to specify what the element was that was rage clicked and maybe a count as to how many times it was clicked within the triggered time.

    Rage clicks can also help you to sniff out areas of your website that are just to slow or need to show the user that something’s happening in the background (a loading gif for example).

    Thanks again!

    I really hope it’s helped and even if you’re not in or working with eCommerce clients, I hope it’s inspired you to think of your next big idea that your customers will love you for.

    Any questions, please get in touch. I’m all ears!

    Looking for the slides?

    Click here to download my slides!

    Cheers!