How (and when) to create a custom metric
Published 2019-01-23, updated 2023-09-15
Summary - Commonly-used metrics do the trick most of the time. But no business is completely like another. Custom metrics go beyond the basic and ensure you cover the specifics that are important to your success.
Business metrics come in all shapes and sizes.
Most of them are ones with which we’re already familiar, such as click through rate, profit or return on ad spend.
But sometimes a pre-existing metric just doesn’t cut it.
By creating a custom metric you can get the measurement you’re looking for without sacrificing the need to accurately measure success.
In this post we’ll look at how to create a custom metric, as well as situations where a custom metric would be best put to use.
What is a metric?
We use business metrics all the time.
But rarely do people who use analytics in their jobs ever stop to ask: Just what is a metric anyway?
Klipfolio defines it as follows: A metric is a quantifiable measure that is used to track and assess the status of a specific process.
A metric can be calculated by simply counting up the number of times something happens. For example: The number of times users have visited your website.
Metrics can also be the product of two or more separate metrics combined into one. Average pages visited per user, for example, is calculated by (you guessed it) totalling up the number of unique page visits on your website and then dividing it into the total number of users.
If you want to make things even more complicated, you can also do additions and subtractions to express a number as a percentage.
Retention rate, a popular metric for human resources professionals, is a good example.
This is calculated by subtracting the number of people who left a company during a given period from the total number of employees at the start of that period, then dividing it into the total number of employees at the start of that period.
Expressed as an equation, it would look something like this:
(Number of employees at the start of a time period - the number of employees who departed)/The number of employees at the start of a time period
So let’s say your inputs look something like this:
- Number of employees at the start of a time period: 100
- Number of employees who departed: 2
That would give you the following equation:
= 0.98 *100
= 98% retention rate
That means that, of the number of people who were with a company at the start of a time period, 98 percent stayed.
The point is this: adding, subtracting and dividing can go a long way towards creating new metrics.
What is a custom metric?
A custom metric is a metric you design to accurately measure an area of interest or monitor specific data.
It can be created using a multitude of means, including:
- Counting the number of times something happens
- Dividing one metric into another
- Subtracting one metric from another then dividing it into another number to express a metric as a percentage
- Amalgamating various data points into a table to provide added context
What makes a metric “custom”? Its use.
If a metric isn’t already extensively in use and you’ve developed it yourself, you’ve got a custom metric – even if it’s just a combination of other metrics that already exist.
The advantage of creating a custom metric
Whatever field you’re in, there are probably some metrics that are useful over and over again.
For the SaaS world, churn rate (the number of subscribers who discontinued their subscription during a period divided into the total number of subscribers) is of constant interest.
For human resources (HR) professionals, measuring retention rate (number of employees who remained with a company over a given period divided by the total number of employees) is important.
Commonly-used metrics serve their purpose. They can help compare your performance to other companies’ through industry standards and ensure a common approach to measurement based on what has worked for others.
But no business is completely like another. There are times when the metrics that dominate your industry or field won’t help you track everything you need to be successful, particularly when you’re using performance measurement tools such as key performance indicators (KPIs) and objectives and key results (OKRs) for setting goals.
Once you start chasing a particular metric for use as a performance measurement tool, you need to be certain that it’s one that will define success for your business.
For example: Click through rate has a lot of uses for digital advertisers, for a few different reasons.
- It helps to compare campaigns that have different budgets: If you were to just measure clicks between one campaign that had an $1,000 budget and one that had an $100,000 budget, you wouldn’t be able to tell which was spending its budget most effectively.
- It tells you how many people are clicking through to your website after seeing your ad: A low click through rate is a sign that people are seeing your ad but choosing not to engage with it.
But what they don’t tell you is how effective you’ve been at getting clicks from the types of people you are trying to reach.
One thousand clicks to a landing page is great, but they don’t tell you a) what they did after they clicked through to the landing page (i.e. did they take the desired action?) or b) if they were the people you were trying to reach in the first place.
A custom metric in this scenario might measure how many women aged 35 (your target demographic) clicked through to your landing page and then signed up for a free trial divided by the total number of people who saw your ad.
That would give you a percentage look at how effectively you are reaching your target audience and ensuring they take the desired action.
It’s the sort of metric that only works with your business, helping to give you a leg up on the competition in matching your performance measurement exactly with your business needs.
When can a custom metric help you?
Sometimes, basic metrics will do the trick for your monitoring needs. But that isn't, and shouldn't, always the case.
Here are the instances where investing the time necessary to develop a custom metric might be worth it
When the current metrics aren’t working for you
Metrics are great ways to get a quick update on how you are performing against your goals.
But each metric is also necessarily only a partial snapshot of the situation.
Take cost per click, another common metric for online advertisers, as an example.
This metric, which tells you how much you are spending on average to get someone to click through to your site, is a good way of determining how far your advertising dollars are going.
But it doesn’t tell you important information, such as how many clicks you are getting and whether you are reaching the right people.
If you need to develop a more complete picture of what’s taking place, then it might be time to add some custom metrics.
When you are working on a special project
For “business as usual”, the standard metrics will frequently suffice.
That won’t always be the case, though, when you’re working on a special project.
If you are working on a one-off campaign or a brand-new project, it might make sense to adopt a custom metric that will work specifically for that project.
When you’re looking to take a deeper dive
Custom metrics don’t need to be used at the expense of more standard ones.
In many cases, custom and standard metrics can be used to form a deadly one-two punch to ensure you’re getting a complete picture of what’s going on.
Standard metrics can help to give you an overview of what’s going on at a company, while custom ones can help give you a deeper dive.
When you work in a new and emerging field or specialized industry
Some people work in fields where there just aren’t that many metrics that will be of use.
So they have no choice but to dive into the field of custom metrics.
Ditto for emerging areas.
If you work in a new and emerging field, then the old metrics might not apply to you.
Take account-based marketing (ABM), for example. Standard marketing metrics, in many cases, won’t apply for account-based marketers.
Why? Because account-based marketers don’t care as much about metrics like traffic volume.
For ABM, the number of people who are interacting with your company is less important than ensuring the right people are interacting with your company.
That’s why metrics like marketing-qualified accounts and reach within an account – both measures of how successfully you are reaching targeted accounts – are better metrics for measuring ABM success.
Of course, many ABM metrics are now considered “standard” in their field. But that’s only because nascent ABM’ers took the time to develop custom ones when they were first starting out.
How to create a custom metric
There are a ton of ways to develop a custom metric – why else would they be known as “custom”?
Here are a few ways you can get started:
Use a specific audience segment
The problem with many common metrics is they don’t take your specific audience segment into account.
With a custom metric you can identify a specific action you want completed from a specific audience.
For example: Maybe you want male users aged 35 and up who use an iPhone to spend more than 30 seconds on a particular page.
You could set this up (relatively) easily as a custom metric in Google Analytics.
That way you’re measuring a specific action for an audience that matters only to you.
Find the metrics that matter most to you, and combine them
What makes click through rate such a useful metric?
It combines two metrics – ensuring your ad is seen and then that users click on it – and expresses it as a percentage.
Consider how you might be able to apply this thinking to developing your own custom metrics.
Maybe a key issue for your company is getting people who have already subscribed to your product to learn about a key feature. In this instance, you might want to divide the number of people who spend a certain amount of time on a page (say, three minutes) into the total number of subscribers who arrive on that page.
Find an area of common measurement
Some metrics are useful because they help to compare two or more things to one another.
Let’s say, for example, we are running two advertising campaigns. One has a budget of $1,000 and the other has a budget of $100.
Using a metric like clicks or impressions isn’t going to be useful because the bigger budget campaign is likely going to blow the smaller budget campaign out of the water, regardless of how effective it is.
That’s where metrics like CTR and CPC help. By expressing the campaign performance in this way, we can determine which used its budget most effectively.
The same thinking can be applied to building custom metrics.
If there are two areas you want to measure which are similar but different in some important ways, you may want to consider creating a custom metric so you can compare him.
To take the example cited above further, maybe you offer a number of different software tools on a subscription basis. One is more expensive to subscribe to than the other, so it makes sense that it would generate fewer sign-ups.
If you wanted to measure how effective each service was in generating sign-ups, you might want to adopt a custom metric that finds an area of common measurement between the two.
Only then can you properly compare them.
How to put your custom metric to work
Setting a custom metric is one thing. But you also need to ensure you start measuring it and, if you choose to adopt it as a KPI or OKR, build it into your strategy.
Some tools, such as Google Analytics, allow you to build custom metrics into the platform so you can regularly review them.
In other cases, you’ll need to calculate it manually.
Custom metrics can help measure your overall strategic success.
If you’ve developed a custom metric that you think would be useful to measure your success, you may want to consider adopting it as a KPI or OKR.
Custom metrics can take some time to develop.
But don’t let that deter you.
If you’re not getting the measurement you need from your standard, run-of-the-mill metrics, perhaps it’s time to consider developing a custom one.
Mark Brownlee is a digital marketing strategist in Ottawa, Canada.
Maximizing Business Insights: The Power of dbt's Semantic Layer with of Klipfolio's PowerMetrics
By Jeroen Visser — November 27th, 2023
Unlocking the Value of Metrics for Everyone: Introducing PowerMetrics + revamped dbt Semantic Layer
By David Mennie — October 17th, 2023
Promoting data literacy with metrichq.org and the power of AI
By Allan Wille, Co-Founder — October 12th, 2023
Let’s fix analytics so we can stop asking you for dashboards
By Cathrin Schneider — September 11th, 2023