How to: Create custom metrics in PowerMetrics

We talk a lot about metrics at Klipfolio. We’re passionate about metrics and how you and your team can use them to drive better business decisions.

Metrics are a core part of PowerMetrics. The platform enables people like you to create metrics, prepare and model data, build dashboards and share insights with your team.

PowerMetrics offers multiple ways to create a metric, broadly categorized as: Instant Data Feed Metrics, Custom Metrics (which include Custom Data Feed Metrics, Data Warehouse Metrics, and Semantic Layer Metrics), and Calculated Metrics.

In this article, I’m going to dive into the various types of Custom Metrics. Be sure to check out this resource for more information on instant metrics and this resource on calculated metrics.

First, let’s start with the basics. 

What is a metric?

A metric is a measurable value that represents performance, success, or progress towards a specific goal or objective. Metrics also store the history of the value over time and can be visualized multiple ways.

For example, let’s say you want to track your Monthly Recurring Revenue—how it changes month over month and which vertical has the highest amount of new subscriptions or expansions. To create an MRR metric, you would take your MRR data from your original data source (like a spreadsheet or accounting software), model your data into a data feed with the dimensions you need, and create a metric using a calculation. Now you have a quantifiable metric that you can use to monitor performance.

Metrics visualize your data, making it easy to analyze and understand versus sifting through a spreadsheet and doing back-of-a-napkin math.

I mentioned earlier that PowerMetrics offers several ways to create a metric. Let’s take a look at the types of Custom Metrics.

What is a custom metric?

Custom metrics are true to their name — a metric built using your own data. This data can come from three main paths, resulting in three different metric types:

1. Custom Data Feed Metrics: Built using a data feed you create and manage from sources like spreadsheets, SQL queries, REST APIs, or account-based services that don't offer an instant metric.
2. Data Warehouse Metrics: Built by querying a data warehouse (like Snowflake or Google BigQuery) directly. Data remains in the warehouse and is visualized in PowerMetrics.
3. Semantic Layer Metrics: Built by querying a dbt Semantic Layer project or Cube directly. Data remains in the source and is visualized in PowerMetrics.

By comparison, Instant Data Feed Metrics use pre-built connectors to third-party services like HubSpot or Stripe, where we auto-model the data for you into a data feed and require no further configuration.

To create any custom metric, you start by connecting your data—whether that's pulling data into a data feed or connecting directly to a data warehouse/semantic layer. Once your data source is set up, PowerMetrics automates the data collection and starts to stack up your data history.

What is a data feed?

For Custom Data Feed Metrics, the most common way to get data into PowerMetrics is by creating a data feed (from sources like APIs, spreadsheets, or uploaded data) which you can prepare in our data feed editor. A data feed is a clean, prepared channel between your data source and the custom metric you create. Data feeds ensure your metric data is standardized, enriched, and ready to be used.

Think of metrics like building blocks. For metrics to fit together in order to compare time periods or dimensions, they need to have the same foundation. A standardized set of metrics gives you and your team a rich analytics system built on trusted data.

A data feed is a cleaner, metric-ready version of your data that is easy and quick to work with. When you model your data in a data feed, you define and manipulate the data you want to include by applying formulas or data formats, like text, number, percentage, currency, date, or duration. Once your data feed is set up, you can create a metric with it. It’s worth noting that a data feed can be used to create multiple metrics—there’s no limitation to how many times you can use your data.

How-to create a custom metric

Let’s look at the steps for creating a Custom Data Feed Metric in detail.

1. Connect your data

In this step, you’ll gather your data, whether it’s an Excel spreadsheet, an SQL query, or data you’ve exported from a cloud application. Ideally, you should try to make sure there is a relevant date with a timestamp that you can add to your data feed, too. Once you have it ready to go, you’re ready to move into the second step: editing your data feed.

Level up data-driven decision making

Make metric analysis easy for everyone

Get Started Now

2. Finalize your data feed for use in your metrics

In this step, you’ll bring your data into our data feed editor. Here you’ll be able to further refine and format your data, like removing null values or fixing text discrepancies. You can also use formulas in the data feed editor to populate additional columns in your data, or merge the feed with another for a richer data set.

When you’re editing your data feed, it’s important to consider the metric you’re planning to create with it. Does your data feed contain the right information, like numeric values you want to track or records to be counted? Are all of the dimensions you want to use in your metric available in your data feed so you can segment your data and look at it from different perspectives? Think about the columns you need, the dimensions you want to filter by, and the time dimension and make sure it’s included in your data feed so we can store it in your metric.

A few other tips to be successful with the data feed editor:

• Don’t over prepare your data. You want to have access to the flexibility and customization available in PowerMetrics.
• Use the unpivot function to convert data from a pivot table format to a list table.
• Use formulas to optimize your data, like grouping it into categories to simplify your metric.
• Group related columns, like first name and last name, into combined dimensions, like a name column, for clarity and ease of use.

3. Create a custom metric

It’s time to create your metric. Here you will select your measure (metric value), segmentation (dimensions), and date and time from the data feed, as well as define the date handling and display settings for your metric.

• Date and Time settings allow you to select the column from your data feed that contains the date/time associated with each metric value.
• Date handling settings will allow you to choose whether to use all values in your data feed for your metric, or to only use the latest values in a given period.
• Display settings is where you name your metric, choose the data format, like numeric, currency, percentage, or duration, and optionally set a favourable trend.

Other features for custom metrics

Configuring data feeds is a core feature for one type of custom metric, but there are a few others that apply to all custom metric types.

Data history and storage

PowerMetrics can store up to 10 years worth of data for you. Data history and the ability to compare metrics across time periods is key to making data-driven business decisions.

Traditional data storage software requires manual updates. Once you configure a metric in PowerMetrics, your metrics will pull in and incorporate your new data automatically, so you can always have the most up-to-date information at your fingertips.

PowerMetrics also offers backfill. Instant metrics have backfill auto-enabled. Backfill is available for custom metrics, but would require you to pull the custom data in from your original data source. And, if your data feed doesn’t have historical data, we start recording it for you as soon as you create a metric with that data.

Query Builder

Query Builder is a powerful tool used to select and connect the tables and fields in your custom data from supported data services. For example, you could connect your Google Analytics 4 data, select the Acquisitions table, and then select columns like New Users, Sessions, and Session Medium to be used in your data feed. You can preview it, add additional columns like a date to give your data a shape, or add optional filters if the API allows it. Then, you can create a Custom Data Feed Metric from this data feed.

It’s a metrics-based approach to reporting and analytics

Metrics are dynamic and enable you to set up your dimensions, aggregation types, history, and trend indicators, giving you complete control from start to finish. You can also add Goals & Notifications to all metrics to monitor your progress, anywhere, anytime. Build, manage, and explore your metrics to match your skill level, technical expertise, or data goals—all in PowerMetrics.