In most of our daily interactions, having many choices is a boon to our quality of life. The ability to choose between products, services, and brands allows us to more often get exactly what we want when the outcome is of high importance. Recently, however, the rise of trends such as minimalism and the star power of ‘organizational consultant’ Marie Kondo could be indicative of the fact that either consciously or subconsciously, many are feeling taxed by the sheer volume of decisions we need to make just to get by on a daily basis.
“Choice overload” is a feeling many of us are familiar with, even if you haven’t heard the specific term. For example, when you’re at the supermarket looking for a frozen pizza, and your eyes glaze over as you browse through the dozens of options, wondering all the while: what ever happened to just pepperoni?
It’s been said that on the internet, “shelf space is infinite.” And while that’s true for e-retailers, who are no longer constrained by how much inventory they can fit into a physical store, that mindset can lead brands to assume that there’s no limit to how much content they can feature on a given site, or individual page. Not interested in the content we’re featuring above the fold? Just scroll or swipe — there’s bound to be something else that’ll catch your eye!
But what if choice overload isn’t just a phenomenon we experience when we’re in a supermarket, or browsing through Netflix? Can websites overwhelm their audiences with too many (well-intentioned) choices of content? Over the years, through a variety of experiences working with various organizations, I started to believe that indeed, choice overload is a factor that saps the performance of websites and apps — often on the very pages and screens to which brands invest the most money in driving traffic.
This concept isn’t one that has only been anecdotally observed; indeed, extensive empirical research supports this paradoxical relationship.
If that were true, it would mean that the simple act of simplifying the experiences we put in front of our audiences could unlock major improvements in performance. So, I set out to help organizations measure whether they’re overwhelming their audiences, and what the performance improvement might look like if they were able to take corrective action.
Measuring “Choice” in Digital Experiences
In order to assess the extent to which choice overload might be reducing the performance of a page, site, app, etc., it’s critical to be able to define what “choice” really means. There’s not any one right answer, so define choice in a way that makes sense in your context. Consider the following two examples: how many choices is Subaru offering on its US homepage?
If we count fairly granulary, we could say there are 21 different choices on this page — and even more than that, really, when we consider items nested within drop-downs. Alternatively, though, we could look at choices more like this.
Here, we’re counting far fewer choices, although the underlying experience is of course the same. When thinking about how to define choice in your environment, consider the following:
- How does your organization think about IA (information architecture)?
- Can you make very granular changes to your experiences, or are you likely to be making broad-strokes changes?
- How flexible is your CMS?
These considerations, among others, will help you determine whether you should be measuring choice in a granular fashion, or if perhaps a more high-level approach will be suitable.
Once you’ve decided what constitutes a choice in your environment, you’ll need to ensure that all of those choices are tracked in the analytics platform of your choice. Either Google Analytics or Adobe Analytics will work fine to support this analysis, although for a variety of reasons we find that the analysis scales more efficiently if you happen to be using Adobe. Regardless, the key is that anything you’d like to count as a choice needs to be tracked — for example, as an event — so that you can determine the quantity of choices on any given page.
Identifying Choices Using Google Analytics
As mentioned above, the prerequisite is that everything you’d like to track as a choice needs to be tracked as an event. Once that’s true, measuring the quantity of choice on a given page is as simple as counting the number of (qualifying) distinct events that correspond to a choice on the page. Note that some events may not correspond to a choice on a page, and hence should be omitted from the count.
Here’s a simple example of how this might look. In Google Analytics, configure a custom report that looks like this.
Note that this report template accomplishes several important tasks:
- First, it gives you full visibility into the events that qualify as choices, because it pulls in category, action, and label-level data. (Note that label-level data may not be necessary in every case.)
- Next, this report template also allows you to exclude any events that don’t correspond to choices on the page. For example, if you’re measuring scroll tracking using events, you’d want to exclude those events, as those events don’t help you understand how much content lives on the page. (As another example, if you use events to measure page load time, you’d want to exclude those events as well.)
- Finally, it allows you to specify the particular page for which you’ll be measuring the number of choices. This is key, and also the primary drawback of doing this analysis in GA. For every page you wish to add to your Choice Overload analysis dataset, you’ll need to manually edit your report template and update the page you’re specifying in the “Page” filter. While this is of course a very easy process to go through, it may not scale well if you have hundreds or even thousands of pages that you’d like to consider.
When you run this report in Google Analytics, you’ll get output that looks something like this.
At first glance, this flat table is nothing but a breakdown of how many events have occurred, for any given combination of event category, action, and label. However, if you look closely, you’ll notice that the metric you care about in this case isn’t actually part of the report dataset itself; rather, what you care about is the number of rows in this table. In this example, that number is 31, which means that there are 31 qualifying events for this page. Put more simply, there are 31 choices on this page.
From here, building out your full dataset is rather straightforward, if somewhat time-consuming. Simply run this report once per page that you care about, and note the number of rows in the resulting table. You can assemble this data in Google Sheets, Excel, or wherever you prefer. You’ll end up with a spreadsheet that looks something like this.
From here, the rest of the analysis process is really up to you. Now that you know how many choices each page in your dataset has, the next steps are as follows:
- Pull in the “outcome metrics” or KPIs that you care about for these pages. Metrics could include clickthrough rate, bounce rate, exit rate, or many others.
- Consider the dimension of the analysis that you’d like to include. If you’re curious about how choice overload might impact users on different devices, for example, include GA’s “Device Type” dimension, and so on.
- Join your choice data with the more standard performance data you pull from GA, using Excel’s V-Lookup function (as one option).
Once you’ve done these steps, you’ll be ready to do the actual analysis itself, exploring the relationship between how many choices are present on a page, and how the page “performs” against your chosen outcomes.