Cut through the Analytics Clutter, Part 1: Why Standard Reporting Isn’t Built for Diagnostics
Published: October 21, 2015
Author: Leslie To
Reporting shouldn’t be just some semi-regular venture where we look into analytics platforms, find performance numbers, dump them into our visualization tool of choice, and then simply hand it over to our bosses or clients. Passively talking about performance and not doing anything about what we see can quickly lead to a stale SEO engagement, one that is far from what we’d like to call “performance marketing.”
Stale SEO plans rely on the same steps, in the same sequence, to optimize the same campaign features. But mindful reporting – which means taking those platforms and numbers and devising intelligent views into the opportunities they reflect for a business – is living, breathing, and critical to unlocking vast revenue potential.
In this three-part series, I’ll focus on Google Analytics – free and available to all! – and examine: the limitations of standard reporting; ways to customize reports to help you dig into truly meaningful data; and segmentation tricks to illuminate performance trends that can chart your course to more profitable actions.
Dissecting Standard Reporting
Most standard reports in Google Analytics come with pre-selected metrics that fall into three general categories: Acquisition, Behavior, and Conversions. Now, you can toggle to whatever dimension you’d like (assuming that we’re dealing with similar scopes), but for the most part, you’re dealing with a core set of metrics.
For my analysis below, I’m going to look at the Default Channel Groupings report:
As far as standard reporting goes, it’s quite useful. You can quickly get a pulse on what’s going on across all acquisition channels and understand which channels struggle with volume, poor high level engagement metrics, or low conversion rates. But that’s really all this report (and a lot of standard reports) can give you: a pulse on what’s going on, but not enough information for you to problem-solve. Standard reporting often lacks the granularity for proper diagnostics.
Let’s get into why diagnostics is difficult with standard reporting.
We have three metrics under the acquisition bucket: Sessions, % New Sessions, and New Users. Sessions is an important metric to establish volume. If there’s a channel or landing page that isn’t driving the volume you need, we have to be able to see it right away. I’ll give this one a pass.
% New Sessions & New Users is where I start to bump into a few problems. So much of SEO is focused on non-brand opportunities and new user acquisition. But this report doesn’t give us the opportunity to understand how these New Users compare to our returning customers. We can get a macro-view of volume, but little else beyond that. For example, some unanswered questions include:
-How do new users act differently than returning customers?
-How are our new customers consuming (or not consuming) our site content or products?
-Are we acquiring these customers are a rate that is acceptable?
-How are we retaining these new users?
-Are these new users converting at a healthy rate?
Behavior varies so much between landing pages and channels, and that variable is usually intent. Assuming that your query-to-intent-to-landing page mapping is accurate, this report lacks the depth you need to fully understand how your users consume or don’t consume your content.
For this particular dimension, again, these metrics aren’t that useful beyond just making sure that these channels aren’t sending users to dead pages.
It’s not the individual metrics that I take issue with as much as the limited fashion in which they are displayed. Different channels, pages, landing pages, etc., will have different intent, and based on that intent, conversion type and conversion rates will differ. Since we are forced to toggle between individual conversions or all conversions in aggregate, we can’t get a good sense of how users move through our conversion funnel and whether or not our micro-conversions are properly set up to accurately measure our customers.
Let’s Wrap It Up & Add a Bow
So while this report has its place, I think it has too much noise for proper diagnostics. Beyond that, it lacks the depth required for performance marketers to truly take action.
The configuration of standard reporting pigeonholes performance based on a number of metrics that can be quite limiting in scope. We lack the ability to control for incremental volume so we have little insight into how efficient our acquisition efforts are. Furthermore, we’re missing sufficient granularity that prevents us from getting into the weeds in a productive way.
We need to get away from pre-selected metrics that inappropriately frame the way we look at performance and start to build reports by first asking ourselves, “What metrics does my client or my business care about most?”
Qualifying reporting with this simple question means that we start reporting at the appropriate place and know that whatever metrics we decide on, they are metrics that reflect key performance indicators. It also allow us to make decisions on whether or not a channel is worth the continued investment, if there are performance gaps we can address with a multi-channel approach, and how we can bucket users into specific categories and create optimizations strategies based on our own data.
Stay tuned for part 2 of this series: reporting customizations that show you performance data that actually means something for your business.