Custom Dimensions and Metrics in GA4

Picture a library where every book has been carefully written, printed, and placed on the shelves, but nobody bothered to add the labels on the spines. The knowledge is all there. You just cannot find anything, because there is no way to sort, filter, or browse. That is roughly what happens in GA4 when you collect rich data but never tell the platform how to organise it. The information exists, locked away, waiting for a label.

Custom dimensions and metrics are those labels. They are the mechanism that turns data you are quietly gathering into data you can actually report on, slice by, and make decisions from. They sound technical, and the official documentation can make them feel daunting, but the underlying idea is simple and you do not need to write a line of code to use them. This guide explains what they are, how they differ, why your custom data sometimes vanishes from reports, and how to set them up with confidence.

Dimensions and metrics: the two halves of every report

Before the word "custom" enters the picture, it helps to be clear on what dimensions and metrics are in the first place, because every analytics report is built from exactly these two ingredients. A dimension is a quality or attribute: the country a visitor came from, the page they landed on, the device they used. Dimensions are usually words. A metric is a number you can count or measure: sessions, conversions, revenue, average engagement time. Metrics are quantities.

Put a dimension and a metric together and you have a sentence the business understands. "Mobile devices" (dimension) generated "forty per cent of revenue" (metric). "The pricing page" (dimension) had "an average engagement time of ninety seconds" (metric). Almost every chart and table you will ever read is some combination of the two. If this framing is new to you, our primer on reading a GA4 report without getting lost walks through how these pieces assemble on screen.

What makes a dimension or metric "custom"

GA4 arrives with dozens of built-in dimensions and metrics, and for many businesses they cover the essentials. But every organisation has questions that the standard options cannot answer because the standard options know nothing about your specific business. A publisher might want to break traffic down by article author. A software company might want to see usage by subscription plan. A shop might want to compare behaviour by customer loyalty tier. None of those exist out of the box, because GA4 has no idea your business has authors, plans, or loyalty tiers.

Custom dimensions and metrics close that gap. They let you take a piece of information you are capturing through an event parameter and promote it into something you can use throughout your reports. A custom dimension carries a descriptive value, like the author's name or the plan type. A custom metric carries a number you want to total or average, like a custom score or a quantity that is not already tracked by default.

The link between parameters and custom dimensions

Here is the relationship that ties the whole system together, and it is the single most important thing to grasp. Custom dimensions and metrics do not collect data themselves. They are merely a way of registering and surfacing data that an event parameter is already sending. The parameter does the collecting; the custom dimension makes it visible. Think of the parameter as the raw ingredient arriving at the kitchen door and the custom dimension as the recipe that lets you serve it.

This is why so many people get stuck. They set up a parameter, see it firing in real time, and then cannot understand why it never appears in a standard report. The answer is almost always the same: the parameter was never registered as a custom dimension, so GA4 collected it but had no instruction to display it.

Understanding scope: event-level versus user-level

When you create a custom dimension, GA4 asks you to choose a scope, and this choice matters more than it first appears. An event-scoped dimension describes a single action. The search term someone typed, the title of a video they played, the category of a product they viewed: these change from event to event, so they belong at event scope. A user-scoped dimension describes the person and sticks with them across their whole relationship with you. Their membership tier, their account type, whether they have ever made a purchase: these describe the human, not the moment.

Getting scope right is the difference between a dimension that behaves sensibly and one that produces baffling numbers. A useful test is to ask whether the value could change several times in one visit. If yes, it is probably event-scoped. If it stays the same all day, it is probably user-scoped. User-scoped dimensions are particularly powerful when you later want to build an audience of a particular kind of customer.

Setting one up, step by simple step

The mechanics are refreshingly straightforward. Inside your GA4 admin area you will find a section called custom definitions. From there you create a new custom dimension, give it a clear display name, pick its scope, and connect it to the event parameter that is already supplying the data. Save it, and from that point on GA4 begins making the value available in reports and explorations. Custom metrics follow the same path, with the additional step of choosing a unit of measurement, such as a standard number or a currency value.

There is one quirk worth knowing. Registration is not retroactive in the standard reports. Data collected before you created the custom dimension generally will not be backfilled, so the sooner you set up the definitions you know you will need, the more history you accumulate. It is a strong argument for planning your tracking thoughtfully at the outset rather than bolting it on after months of regret.

Where the parameters come from in the first place

Custom dimensions need a parameter to point at, and those parameters usually arrive through your tagging setup. For many teams that means a tag management system, which lets marketers create and send custom parameters without editing site code each time. If you have not set that up yet, our guide on how to set up Google Tag Manager is the natural starting point, since it is the layer where most custom data is born before it ever reaches a custom dimension.

Watch the limits and keep things tidy

GA4 does not let you create unlimited custom dimensions and metrics. There are caps, and while they are generous enough for most businesses, they reward discipline. Every custom definition you create should earn its place by answering a real question you will revisit. Creating dimensions speculatively, just in case, clutters your account and uses up your allowance. A short, well-chosen set beats a sprawling collection nobody understands.

Tidiness pays dividends in another way too. Clear, consistent naming means anyone on your team can open the reports and immediately understand what each dimension represents. Vague names like "dimension1" or "custom_thing" become a mystery within weeks. Treat your custom definitions like a well-kept index, and the whole account stays navigable as it grows.

Putting custom data to work

Once your custom dimensions are live, the fun begins. You can add them as a breakdown in standard reports, drag them into a flexible exploration to study them from every angle, and use them to define precisely the audience you want to understand or reach. A custom dimension recording subscription plan, for instance, lets you compare conversion behaviour across plans, then build a remarketing audience of the plan most likely to upgrade.

This is also where custom data connects to the bottom line. By layering your own business attributes onto standard behaviour, you can answer questions that directly inform strategy: which customer segment is most valuable, which content keeps people coming back, which product category quietly drives the most revenue. Our overview of the key metrics every business should track monthly pairs well here, because custom dimensions let you cut those headline metrics by the segments that matter most to you. For shops specifically, combining custom dimensions with e-commerce analytics turns generic sales figures into segmented, actionable insight.

A foundation for everything that follows

It is worth seeing custom dimensions not as an advanced extra but as a foundation. They are the bridge between the raw data your site generates and the meaningful, business-shaped reports you actually want. Skipping them leaves you reading GA4's generic view of the world; embracing them lets you read your world. If the whole field still feels unfamiliar, the gentle introduction in what web analytics is and why it matters is a calm place to build your bearings before going deeper.

Frequently asked questions

References

1. Google. "[GA4] Custom dimensions and metrics." support.google.com/analytics.
2. Google. "[GA4] About custom definitions." support.google.com/analytics.
3. Simo Ahava. "Custom dimensions in Google Analytics 4." simoahava.com.