What are metric dimensions?
Overview
Metric dimensions let you break down a metric by any categorical attribute in your data — like region, channel, brand, or market. Instead of creating a separate metric for each segment, you configure dimensions once on a metric source and create filtered metric cards from it.
This article explains what dimensions are, how they work, and the current limitations to know about.
What is a dimension?
A dimension is a categorical column in your metric data that adds context to the numeric value. For example, a CSAT dataset doesn't just have a score and a date — it might also have columns for Country, Booking Channel, or Brand. Each of those is a potential dimension.
When you configure a dimension on a metric source, it becomes an interactive filter. You can then create metric cards that show data for a specific value — like "CSAT — France" or "Signups — Mobile" — and reuse them across any journey in your workspace.
Why dimensions matter
Before dimensions, comparing variations of the same metric required creating a separate metric card for each segment — one for each country, channel, or brand. This caused metric libraries to grow quickly, made maintenance painful, and made comparison across segments effectively impossible inside TheyDo.
With dimensions, you only need one metric source to bring in all variants at once.
Before you start
Already have metrics set up? Dimensions can only be added when a metric source is first created. If you have existing metrics that were set up before dimensions were available, you'll need to reset the metric and recreate it to configure dimensions. Reach out to your workspace admin if you're unsure how to do this.
How dimensions work
- When creating a metric source, designate up to 5 columns in your data as dimensions.
- Once the source is saved, those dimensions are available as filters in the Metrics library.
- In the Metrics library, click the arrow next to the metric source to expand it.
- Click Create metric card and configure the dimension value, time range, and interval (e.g., Country = France, Monthly, Last 12 months).
- Name the card (e.g., "CSAT — France") and save.
- The metric card appears nested under the source and is ready to drag into any journey.
Where dimensions are available
Dimensions work across all metric sources:
- Data warehouses: BigQuery, Snowflake, Databricks
- Amazon S3
- Qualtrics
- Medallia
- CSV import
Good to know
- Up to 5 dimensions per metric source — consistent across all sources. This doesn't limit how many dimensions you can track overall, just how many you can configure on a single metric source. If your data has more than 5 dimensions, split them across multiple metric sources for the same metric. For example, if you're tracking conversions with 11 dimensions, create three metric sources:
Ideally, group the dimensions you most commonly compare together in the same metric source. - Conversions: product, area, domain, country, city - Conversions: state, product line, category, customer segment, brand - Conversions: loyalty segment
- Dimensions must be configured at creation time. Existing metrics set up without dimensions need to be reset and recreated to use them.
- One dimension value at a time. You can filter to a specific value (e.g., France), but you can't overlay multiple dimension values on the same chart yet.