Marketing metrics recipes

This article provides recipes to recreate common marketing metrics, such as session totals, bounce rate, and page views, using Event Segmentation or User Sessions charts. Each recipe lists the module setup you need to build one metric.

Before you start, read the basics of building charts in Amplitude. Some recipes in this article use custom formulas, which the custom formula metrics, syntax, and definitions article explains.

Metric recipes

The examples in this section provide a starting point for setting up each metric. As with other analyses, use filters, group-bys, or segmentation to refine results.

Session totals

The session totals metric provides a sum of sessions.

Use the following setup for each module in your User Sessions chart to create the session totals metric:

• Sessions: Count sessions for all sessions.

  To filter results by a specific domain or URL:

  • Click Filter by and choose Contains Event. Search for the Page Viewed event. Set the count to greater than or equal to one.
  • Click the Filter to add a where clause for your Page URL.

• Measured as: Total Sessions.

• Segmented by: All users.

Visitors

The visitors metric counts unique users who trigger a specific event.

Use the following setup for each module in your Event Segmentation chart to create the visitors metric:

• Events: Page View.

  To filter results by a specific domain or URL:

  • Click Filter by and choose Page URL. Select the checkbox next to your domain and click Apply.

• Measured as: Uniques.

• Segmented by: All users.

Sessions per user

The sessions per user metric gives an average count of sessions by unique user.

Use the following setup for each module in your User Sessions chart to create the sessions per user metric:

• Sessions: Count sessions for all sessions.

To filter results by a specific domain or URL:

• Click Filter by and choose Contains Event. Search for the Page Viewed event. Set the count to greater than or equal to one.

  • Click Filter to add a where clause for your Page URL.

• Measured as: Avg Per User.

• Segmented by: All users.

Average session duration

The average session duration metric gives an average length of sessions.

Use the following setup for each module in your User Sessions chart to create the average session duration metric:

• Sessions: Count sessions for all sessions.

To filter results by a specific domain or URL:

• Click Filter by and choose Contains Event. Search for your Page View event. Set the count to greater than or equal to one.

  • Click Filter to add a where clause for your Page URL.

• Measured as: Avg length (counting in days, hours, minutes).

• Segmented by: All users.

Page views

The page views metric gives a sum of page views events.

Use the following setup for each module in your Event Segmentation chart to create the page views metric:

• Events: Page View.

To filter results by a specific domain or URL:

• Click Filter by and choose Page URL. Select the checkbox next to your domain and click Apply.

• Measured as: Event Totals.

• Segmented by: All users.

Page views per session

The page views per session metric gives an average count of Page View events per session.

Use the following setup for each module in your User Sessions chart to create the page views per session metric:

• Sessions: Count events performed within sessions for all sessions.
  • By Event Count of the Page View event.

To filter results by a specific domain or URL:

• Click Filter by and choose Contains Event. Search for the Page Viewed event. Set the count to greater than or equal to one.

  • Click Filter to add a where clause for your Page URL.

• Measured as: Avg events per session.

• Segmented by: All users.

Bounce rate

The bounce rate metric compares the count of sessions with a single Page View event to the total number of sessions, as a percentage.

Use the following setup for each module in your User Sessions chart to create the bounce rate metric:

• Sessions: Count sessions.
  • A: All sessions.
    • where Contains Event = Page Viewed with a count of one.
  • B: All sessions.

To filter results by a specific domain or URL of each session:

• Click Filter by and choose Contains Event for session B. Search for the Page Viewed event. Set the count to greater than or equal to one for session B.

  • Click Filter to add a where clause for your Page URL for sessions A and B.

• Measured as: Formula %:SESSIONTOTALS(A)/SESSIONTOTALS(B).

• Segmented by: All users.

This User Sessions chart calculates the 30-day bounce rate for the Page URL ampli.com/home. The bounce rate for March 4th was 97.3 percent.

The chart plots the daily bounce rate for ampli.com/home over 30 days, where bounce rate is the share of sessions that contain exactly one Page View. The line stays in the low-to-high 90s all month and peaks at 97.3 percent on March 4, so almost every session that reaches this page views only that page and leaves. A rate this high and this flat flags a landing page worth investigating. The dashed tail marks the most recent day, whose data is still provisional.

Entry rate

The entry rate metric compares sessions with a Page View event grouped by the first property value of the specified event property to the total number of sessions, as a percentage.

Use the following setup for each module in your User Sessions chart to create the entry rate metric:

• Sessions: Count sessions.
  • A: All sessions.
    • Grouped by the First Property Value of the Page View event property Page Title.
  • B: All sessions.

To filter results by a specific domain or URL of sessions A and B:

• Click Filter by and choose Contains Event. Search for the Page Viewed event. Set the count to greater than or equal to one.

  • Click Filter to add a where clause for your Page URL.

• Measured as: Formula %:SESSIONTOTALS(A)/SESSIONTOTALS(B).

• Segmented by: All users.

This User Sessions chart calculates the 30-day entry rate for the Page URL ampli.com/home. The chart's group-by property value Page Title appears as different colored lines in the chart. The entry rate for March 3rd was 35.5 percent for the Home property value.

Each line shows the entry rate for one page title, meaning the share of sessions that begin on that page. Home leads throughout at roughly 30 to 40 percent, reaching 35.5 percent on March 3, while the other five page titles sit well below it. Read together, the lines confirm that Home is where most sessions start.

Exit rate

The exit rate metric compares sessions with a Page View event grouped by the last property value of the specified event property to the total number of sessions, as a percentage.

Use the following setup for each module in your User Sessions chart to create the exit rate metric:

• Sessions: Count sessions.
  • A: All sessions.
    • Grouped by the Last Property Value of the Page View event property Page Title.
  • B: All sessions.

To filter results by a specific domain or URL of sessions A and B:

• Click Filter by and choose Contains Event. Search for the Page Viewed event. Set the count to greater than or equal to one.

  • Click Filter to add a where clause for your Page URL.

• Measured as: Formula %:SESSIONTOTALS(A)/SESSIONTOTALS(B).

• Segmented by: All users.

This User Sessions chart calculates the 30-day exit rate for the Page URL ampli.com/home. The chart's group-by property value Page Title appears as different colored lines in the chart. The exit rate for March 4th was 16.1 percent for the Log In property value.

Each line shows the exit rate for one page title, meaning the share of sessions that end on that page. Home carries the highest exit share, while Log In sits around 15 to 18 percent and reaches 16.1 percent on March 4. Compare a page's entry rate against its exit rate to see where sessions start versus where they drop off: a page that ranks high in both, like Home, both opens and closes many sessions.

Performance marketing metrics

Amplitude supports several common performance marketing metrics. Create these metrics by following the recipes in this section.

Ad network clicks

• Metric type: Formula.
• Event: Daily ad metric.
• Grouped by: ad_metrics.clicks.
• Formula: PROPSUM(A).

Ad network impressions

• Metric type: Formula.
• Event: Daily ad metric.
• Grouped by: ad_metrics.impressions.
• Formula: PROPSUM(A).

Ad network costs

• Metric type: Formula.
• Event: Daily ad metric.
• Grouped by: ad_metrics.costs.
• Formula: PROPSUM(A).

Return on ad spending

• Metric type: Formula.
• Event A: Complete purchase.
  • Grouped by: Revenue.
• Event B: Daily ad metric.
  • Grouped by: ad_metrics.cost.
• Formula: %: PROPSUM(A) / PROPSUM(B).

Customer acquisition cost

• Metric type: Formula.
• Event A: Daily ad metric.
  • Grouped by: ad_metrics.cost.
• Event B: User Sign Up.
• Formula: $: PROPSUM(A) / TOTALS(B).

Clickthrough rate

• Metric type: Formula.
• Event A: Daily ad metric.
  • Grouped by: ad_metrics.impressions.
• Event B: Daily ad metric.
  • Grouped by: ad_metrics.clicks.
• Formula: %: PROPSUM(A) / PROPSUM(B).