Marketing Metrics Explained: Exploring Metric Types and Formats
Header image by @dawson2406 via Unsplash
A metric is simply a unit of measure which helps you quantify something. When it comes to marketing and communications, there are generally five different categories of metrics that serve different purposes and which can help your organize your KPIs (see my other article, How to Categorize Your Marketing Metrics). However, in order to ensure you choose the right metrics to quantify performance, you also need to understand the different types and formats of metrics, what they tell you, and when it's appropriate to choose one metric type over another.
When it comes to selecting meaningful marketing metrics, you need to consider two aspects; 1) the metric type, and 2) the metric format.
The 3 types of marketing metrics
There are three types of metrics you’ll come across in the world of marketing. Some of these metrics come off-the-shelf from your favourite analytics tool, while others need to be engineered. Here’s a breakdown of the three metric types.
Platform Metrics
Aka ‘off-the-shelf’ metrics, these metrics which natively supported within an analytics tool and are easy to access available. Think of metrics like Reach from Facebook or Bounce Rate from Google Analytics. Platform metrics are the majority of metrics that you will use. However, as you’ll see later on, you will sometimes need to create new custom metrics that are more aligned with your business objective.
There’s a helpful principle known as the Streetlight Effect, which is “a type of observational bias that occurs when people only search for something where it is easiest to look.” The issue with platform metrics is that, because they’re so convenient we sometimes use a metric that’s easy to retrieve over something that may be a better measure of campaign performance.
This isn’t too say you shouldn’t use platform metrics. In fact, most of your metrics will usually be platform-based. But, when you’re selecting metrics to measure success, ask yourself whether the platform metrics you have access to really are fit for accurately tracking the performance of your campaign.
Here are a few examples of platform metrics:
Impressions (Twitter)
Reach (Facebook)
Sessions (Google Analytics)
Bounce Rate (Google Analytics)
Derived Metrics
Derived metrics are those you create by combining two or more existing metrics. The existing metrics can either be a platform metric, or something else like another derived metric.
An example of a derived metric is engagement rate. While some social networks do provide native support for an engagement rate, many do not. And so, you may need to calculate engagement rate manually. Facebook is a good example, as it doesn’t provide a platform metric for engagement rate. However, Facebook do provide a metric called “People Engaged”, which can be used to divide into Reach, hence giving you a derived metric (note. since publishing this article Facebook has removed the People Engaged metric).
Another common example of a derived metric is a conversion rate. Although many ad platforms, like Google or LinkedIn Ads do provide a form of a conversion rate, most often you will need to rely on a derived metric to calculate cross-channel conversion. For example, say you want to know the conversion rate of made a purchase on your website out of the people who clicked on an ad. This will require you to blend data from two different data sources (e.g. ad clicks + website transactions), which will ultimately give you a derived metric for conversion.
Here are a few examples of derived metrics:
Ad to website conversion rate (manually calculated by dividing website users with a transaction into unique clicks on your ad)
Facebook engagement rate (manually calculated by dividing people engaged into reach, or total engagements into total impressions)
Engineered Metrics
The third and final type of metric is known as an engineered metric. These are the most difficult metrics to acquire, because they need to be created from scratch. However, they can also be highly valuable, as they’re engineered to meet your specific needs.
An example of an engineered metric is the creation of a custom event that you track on your website. Tools like Google Analytics 4 (GA4) have certain limitations when it comes to things it can quantify out-of-the-box, and many website owners will customize their website implementation to track the things they need. In many cases, they’ll use a tag management solution, such as Google Tag Manager (GTM), to create custom events that aren’t tracked by default on your website.
One example of this is average page load time for Google Analytics. This is one interesting, because this metric was a platform metric in Universal Analytics (UA). But when Google upgraded to GA4, native support for average load times went away. And at the time of writing, the only way to get it back is to engineer the metric, which can be done using GTM.
Here are a few examples of engineered metrics:
Page load time in GA4
Custom button clicks in GA4
Custom interactions in your mobile app
5 common metric formats
Beyond the types of metrics, you should also consider the metric format. Here are the five most common formats for metrics that you will encounter:
Numbers (e.g. 50 website sessions)
Percentages (e.g. 2% conversion rate)
Duration (e.g. 2m 35s)
Currency ($50 per transaction)
Ratio (2:1)
Conclusion
Thinking about metrics in terms of the type and format can help you choose metric that are better aligned with your business objectives. This is because this pushes you to think about measurement in a more comprehensive and purposeful way.
If you’d like to learn more about how to tie metric types and formats into a broader measurement framework, check out my Marketing Metrics Planner template, which is available for free on Miro.
