Calculated Metrics

Calculated metrics derive new metrics from existing ones using formulas. They do not require additional data collection — they compute from metrics already in GA4 using simple arithmetic operations.

The most common use cases:

• Revenue per user: purchase_revenue / total_users
• Conversion rate for a specific event: event_count / sessions
• Average order value: purchase_revenue / transactions
• Cart abandonment rate: 1 - (transactions / add_to_cart_events)
• Cost per acquisition: requires custom metric for cost data

Limits

Tier	Calculated metrics limit
Free GA4	5 per property
GA4 360	50 per property

The 5-metric limit for free properties is tight. Choose carefully — focus on metrics that stakeholders regularly need in the GA4 UI and that cannot easily be computed manually. For complex or ad-hoc calculations, BigQuery is more flexible.

Creating a calculated metric

1. Go to Admin → Custom definitions → Calculated metrics tab.

2. Click Create calculated metric.

3. Enter a Metric name — this is what appears in reports. Use clear, descriptive names: “Revenue per User”, “Lead Conversion Rate”.

4. Enter a Description — document the formula and purpose.

5. Select a Unit:

   • Standard: plain number
   • Currency: formatted as currency
   • Feet/Miles/Meters/Kilometers: distance
   • Milliseconds/Seconds/Minutes/Hours: time
   • Percent: displayed as a percentage

6. Build the Formula using the formula editor. Click metrics to add them, use arithmetic operators.

7. Click Save.

Formula syntax

Formulas use:

• Metrics: click to insert from the picker, or type {{metric_name}}
• Arithmetic operators: +, -, *, /
• Parentheses: for order of operations

Examples:

# Revenue per session
{{purchaseRevenue}} / {{sessions}}

# Conversion rate (any conversion event)
{{conversions}} / {{sessions}}

# Average order value
{{purchaseRevenue}} / {{transactions}}

# Engagement rate (already built-in, but as example)
{{engagedSessions}} / {{sessions}}

# Revenue per user
{{purchaseRevenue}} / {{totalUsers}}

Caution

GA4 will not warn you if your formula produces division-by-zero for some dimension combinations. When a denominator metric is 0 for a row (e.g., a page that had sessions but no purchases), GA4 typically shows 0 or “(not set)” rather than displaying an error. Verify calculated metrics across different filter contexts to understand how they behave with edge cases.

Practical examples

Revenue per user by channel

Formula: {{purchaseRevenue}} / {{totalUsers}} Unit: Currency Use case: Identify which acquisition channels bring users with the highest average lifetime revenue in the analysis period.

Create this metric, then view the Traffic acquisition report with “Revenue per User” as a metric column. Compare Organic Search vs. Paid Search vs. Email revenue efficiency.

Lead conversion rate

Formula: {{conversions:lead_form_submission}} / {{sessions}}

Wait — GA4 does not let you filter by event name within a calculated metric formula. You cannot reference a specific conversion event by name in the formula. The {{conversions}} metric represents all conversions.

Workaround: Create a custom metric for lead submissions (if you push a numeric value with lead_form_submission), or use Explorations to filter by event name.

This is a real limitation of calculated metrics — they operate on aggregate metrics, not filtered slices. For per-event conversion rates, use Explorations with event-level segmentation.

Cart abandonment rate

Formula: 1 - ({{transactions}} / {{addToCartEvents}})

Note: addToCartEvents is an automatically collected event count. The formula shows the proportion of add-to-cart actions that did not result in a transaction.

Unit: Percent

Cost-efficiency metric

If you send ad spend as a custom metric (via the Measurement Protocol or a cost data import), you can compute:

Formula: {{customMetric:ad_spend}} / {{conversions}} Unit: Currency Label: Cost per Conversion

Using calculated metrics in reports

Calculated metrics appear in the same places as regular metrics:

• Standard reports: add as a metric column in a customized report
• Explorations: available in the metrics picker
• Comparisons: usable in comparison metrics

They are not available in:

• Audiences (you cannot filter audiences by calculated metric values)
• Segments (same limitation)
• Google Ads bidding (Smart Bidding uses native conversions, not calculated metrics)

Calculated metrics vs. BigQuery

Calculated metrics cover simple arithmetic on GA4 aggregate metrics. For anything more complex, use BigQuery:

Need	Calculated Metrics	BigQuery
revenue / users	Yes	Yes
Revenue percentile analysis	No	Yes
Attribution-adjusted conversion rate	No	Yes
Cohort-based LTV calculation	No	Yes
Moving averages	No	Yes
Ratios with event-level filters	No	Yes

If your stakeholders need it in the GA4 UI and it is a simple ratio of existing metrics, use Calculated Metrics. For everything else, build it in BigQuery and connect to Looker Studio.

Common mistakes

Trying to filter within a calculated metric formula

You cannot write {{conversions WHERE event_name = 'purchase'}} in a formula. The formula editor only accepts metric names and operators. If you need a rate for a specific event, the alternatives are:

1. Create a custom metric for that event’s count (requires sending a custom numeric parameter)
2. Use an Exploration with event name filtering
3. Compute it in BigQuery

Using the wrong denominator

{{purchaseRevenue}} / {{sessions}} and {{purchaseRevenue}} / {{totalUsers}} tell very different stories. Sessions can be much higher than users for repeat visitors. “Revenue per session” rewards high-frequency visitors; “revenue per user” normalizes by person. Know which question you are answering.

Wasting calculated metric slots on metrics easily computed manually

If a stakeholder needs “revenue per 1000 users” once a quarter and can do the math from existing report numbers, that is not worth a calculated metric slot. Reserve the 5 slots for metrics that appear in regular reports and dashboards.

Related articles

Custom Metrics Register numeric event parameters for aggregation — the complement to calculated metrics.

Explore Reports Use explorations for complex metric analysis including event-level filtering.

Common Queries Compute any ratio or derived metric you need in BigQuery.