Cut through the Clutter, Part 2: The Art of Creating a Custom Report
Published: October 22, 2015
Author: Leslie To
When we were last together, I dissected standard reporting and introduced the idea of asking yourself a key question before we start reporting. And it is with this question that we begin today’s conversation: what metrics & dimensions do my client or my business care about most?
Metric selection is the first, arguably the most important, step in reporting. If we are not mindful in metric selection, we risk running into the same problems we saw with standard reporting.
Now, if any of you have had experience with custom reporting in Google Analytics, you’re well aware that there are a lot of metrics to pick from. And I will not preclude you from picking whatever metrics you like, as long as they matter to your core business. But I do think that there are just some metrics that are better than others…
Sidebar: I want to take a quick break to square up some definitions. For those unfamiliar with Google’s terminology, a metric is any number or ratio (e.g. sessions, bounce rate, average visit duration, etc.), whereas a dimension is information about your user (e.g. channel, device type, landing page, etc.). Understanding the difference is key, especially when we’re talking about custom reporting. You should also have a strong grasp of reporting scopes within Google Analytics. For those unfamiliar, I recommend reading this really amazing piece by Avinash Kaushik; in it, he breaks down the difference between hits, sessions, and user-level dimensions and metrics. If you’re looking for a technical breakdown of scope, feel free to peruse this article, in which Google explains the importance of scope for custom dimensions (the basic premise applies to custom reporting as well).
Pro Tip: Build a Report to Answer Business Questions
Mindful reporting should endeavor to answer three basic inquiries:
1. Are my marketing channels working the way I expect them to?
2. If not, where are my performance gaps?
3. And what can I do to address those performance gaps?
Our custom report will try to answer question #1, and segmentation (part 3 of this series…be patient!) will answer questions #2 and #3.
But first… let’s build this hyped report of mine.
Focus on Key Performance Indicators
While you can pick whatever metrics you want, there are a number of metrics (combined with two dimensions) that I really like to look at when I am trying to diagnose performance.
Users & Sessions
These two metrics help establish volume, which is imperative if you have a channel struggling to drive volume. Diagnosing volume issues is a first step to figuring out if your acquisition channels are performing the way you’d expect.
I like to prioritize users over sessions, so I made users the primary metric for this report. At the end of the day, I care more about the number of customers who come to my website rather than the number of times they visit. That is not to say that I don’t care about repeat visits, but my primary objective is to acquire new customers, so this report must reflect that end goal.
This metric helps me understand whether certain channels are better at customer retention than others – a point that we’ll come back to in Part 3, when I talk about the power of segmentation.
Revenue and Revenue/User
The macro conversion for every business is going to be revenue, so we can’t have a report that looks at performance that doesn’t include it. Revenue, however, can be misleading because there can be many factors at play.
We want to control for incremental volume by also looking revenue/user. Essentially, I want to see if and how certain channels grew revenue and at what price. If drove 50% more users but only grew revenue by 10%, then we know our investment wasn’t as effective.
Sidebar: If your Google Analytics account doesn’t have revenue tracking in place, I recommend adding a Goal Value to your goals so that you have a nominal understanding of your acquisition channels. Attaching some kind of number to your goals makes this analysis a lot easier.
Sprinkle in Some Micro-Conversions
I recommend picking micro-conversions that represent users throughout the conversion funnel. If we only measure performance against the macro, we’ll miss out on users who take a bit longer to make a purchase or convert at the macro level. To fully understand performance, we need to know if and how certain channels drive specific conversion types compared to others.
For this particular client, I chose Add to Cart, Registrations, and Wishlist Adds – with Add to Cart as the final micro-conversion point prior to the macro, Registration as the midpoint, and Wishlists as aspirational users who may require lengthier customer nurturing campaigns (or bigger motivations before buying, perhaps in the form of discounts).
Dimensions Are All about Context!
Metrics need dimensions so that they are properly contextualized! I really like the Default Channels Grouping as my primary dimension. First and foremost, organic traffic has never existed and will never exist in a vacuum. We cannot create reports that make that assumption. We need to understand how organic performs in relation to its sister/brother acquisition channels.
Landing page is my preference for secondary dimension drilldown because no site reporting is complete without a landing page report. We want to understand 1) if certain sections of the site are not driving volume; 2) if certain sections are better at converting users in some channels compared to others; and 3) if there are certain sections that are simply not consumed and never convert users.
The Fruits of Our Labor
Now we have a report that not only has more depth and controls for incrementality but is also built specifically for diagnostics.
That being said, we’re still dealing with a fair bit of noise…
Stay tuned for Part 3, where I’ll cover how segmentation can control for noise and make this report actionable!