Building Custom Dimensions and Metrics for Your Analytics

September 27, 2017 Matthew Maddox

Tools-and-papers-with-sketches-1

Because data requirements will be different in every organization, oftentimes the out-of-the-box dimensions and metrics in your analytics tool won’t meet your business needs. Pageviews will only get you so far in understanding your customers.

To get the most out of your analytics tool, you will need to use custom dimensions and metrics.

Building Business Requirements and Selecting KPIs

In a recent article I published about starting with the “why” of your analytics implementation, I talked about the importance of outlining your business objectives before ever touching your analytics tool. The same applies when creating custom dimensions and metrics.

The process of building your business requirements and corresponding KPIs should go something like this:

  1. Identify your organization’s key objectives for your target audience (Objective).
  2. Describe how you hope to accomplish those objectives via the website (Key Business Requirements).
  3. Define how to measure those requirements (Key Performance Indicators).

Here is a simplified example of what this process might look like for a media & entertainment website that generates revenue from advertising:

  1. Objective: Increase the number of videos consumed per user in North America.
  2. KBR: Autoplay appropriate content at the end of each video.
  3. KPI: Average number of videos viewed per user per session (if you successfully target content to your visitor, this KPI should increase).

Your KPIs will help you quantify how well your KBRs are helping to accomplish your Objectives. Once you have defined the KPIs, you can set specific goals to measure if you’re doing better (or not!).

Quick Refresher: Dimensions vs. Metrics

As an analytics trainer, I often found there was confusion about the difference between dimensions and metrics, making it hard to create custom dimensions and metrics.

So here’s a quick refresher to help you know the difference between a dimension and a metric:

A metric is a measurement.

A dimension is a description.

Metrics

All measurements are expressed as numbers. Let’s think about it in terms of athletes and Olympic events. Speed, weight, age and number of gold medals are all metrics because they can be expressed as numbers:

  • Time: 9.63 (seconds to run the 100m: Usain Bolt, 2012 London Olympics)
  • Weight: 130 (kilograms in wrestling: Rulon Gardner, 2000 Sydney Olympics)
  • Age: 15 (youngest age to win a gold medal: Tara Lipinski, 1998 Olympics)
  • Gold medals won: 7 (gold medals, British Cycling Team, 2012 Olympics)

Anything that can be expressed as a number is a measurement or metric. In web analytics, the following are examples of metrics:

  • Visitors
  • Visits
  • Page views
  • Campaign clickthroughs

Each of these are metrics because they are measurable.

Dimensions

Dimensions, on the other hand, are used to describe the who, what, where and when of things like the user or web page, and are used to break down your customer base into segments. Going back to the Olympics, we can describe and put them into context with dimensions:

  • Athletes: Usain Bolt, Rulon Gardner, Tara Lipinski, Philip Hindes
  • Participating countries: Jamaica, USA, Great Britain
  • Medal types: gold, silver, bronze
  • Events: 100 meters, Roman-Greco wrestling, figure skating, men’s team sprint

In the web world, dimensions are similar, they describe the traffic on your website:

  • Country
  • Site section
  • Campaign name
  • Product

Dimensions are ways to describe data by slicing and dicing into smaller pieces so you can make sense of the data (also known as segmentation). So instead of just looking at all of our data undifferentiated, we can compare, for example, the conversion rate between European and North American visitors.

Custom Dimensions and Metrics

When creating custom metrics, you’ll take KBRs and turn them into KPIs. Remember, KPIs are metrics to measure how effective your KBRs are at reaching your Objectives.

Common Custom Metrics

Metrics are often expressed as ratios. This allows us to put them more in context for better comparisons.

For example, the metric “total revenue” is helpful, and revenue might go up as we increase visitors to our site, but looking only at that number doesn’t help us understand the actual efficiency of our site. For that we would need a ratio, like revenue per visitor.

Ratios are a quick way to measure how effective our metrics are over time. (See Custom Calculated Metrics for more information.)

This table shows some common metrics (KPIs) you might create given the following business requirements:

Objective/Business Requirements Key Performance Indicators
Optimize sales efficiency Revenue per visitor
Revenue per transaction
Make it easy to find help articles Searches per article read

Reading time (time spent on article)

Increase advertising effectiveness Clickthrough rate (by campaign)

Return on Ad Spend (by campaign)

Drive advertising on video pages Video percentage watched

Number of video starts

Videos viewed per visitor per session

Custom metrics like these provide a foundation of numbers to help us see how well our site is doing.

Custom Dimensions

Custom dimensions, on the other hand, describe things beyond the dimensions already included in your analytics tool. For example, your tool might include a geographic dimension that automatically looks up the location where the IP address is registered. But figuring out the city is far less reliable than figuring out the country. So you might create a custom “City” dimension and find a more reliable way to record it, perhaps by capturing it via a registration form, storing it, and re-using it each time the visitor returns.

Another example: set up your implementation to track the login state of users. This would allow you to see differences in behavior between anonymous and registered users.

Finally, consider creating custom dimensions that describe elements of your site: site section and sub-section, categories for products or articles and page language.

The more ways you can describe your numbers with custom dimensions, the better you will be at measuring conversion effectiveness among different types of people, interests or site attributes.

Custom Calculated Metrics

You’ll notice many of the metrics in the chart above (like videos viewed per visitor per session) are expressed as ratios. Because metrics are numbers, you can multiply, divide and put them into mathematical formulas to come up with meaningful ratios.

You can create calculated metrics directly within most analytics tools. The process will vary from tool to tool, but here are some resources you can take a look at for Google Universal Analytics and Adobe Analytics:

Universal Analytics: About calculated metrics

Adobe Analytics: Calculated and Advanced Calculated (Derived) Metrics

The Benefits of Customization

Custom dimensions and metrics to match your business requirements and KPIs make data specific to your needs. And while it adds another level of complexity to your site, it has the benefit of helping you measure exactly how well your site is meeting your objectives.

An easy way to validate that your custom dimensions and metrics are being captured reliably is by using automated, rules-based data validation. Check out our blog post Using ObservePoint’s Rules Engine for Robust Tag Validation to learn more.

 

About the Author

Matthew Maddox

Matt’s mission is to educate and enable customers to use the marketing technologies they select for their sites most effectively. Matt delivered training at Omniture and Adobe for over eight years before joining ObservePoint. He was the dedicated trainer for several global companies, creating and delivering custom courses based on their corporate business requirements. With a wealth of experience solving analytics questions in many industry verticals, including e-commerce, media, finance, lead generation and automotive, Matt offers sound direction and analytics insight.

LinkedIn More Content by Matthew Maddox
Previous Article
Find Money-Making Insights in Your Data
Find Money-Making Insights in Your Data

This article discusses how companies can combine multiple data sources to unlock the entire customer story ...

Next Item
Why You Should Audit Your Web Tags
Why You Should Audit Your Web Tags

Why You Should Audit Your Web Tags outlines key considerations surrounding broken tags and how to avoid them.