Interpret your User Sessions chart
Amplitude Academy
Uncover How Long Users Spend in Your Product with User Sessions
Learn how to use the User Sessions chart to analyze the time users spend in your product.
Get startedThe User Sessions chart measures how frequently users start sessions, how long those sessions last, and how those metrics differ by user segment.
Interpret your User Sessions chart
The results in User Sessions depend on the metrics you chose while building your chart, such as a count of sessions or a count of events.
The example below displays a daily count of sessions with at least one Add to Cart event for all users over the last 30 days. The chart also filters users by Country (United Kingdom).
| Date | United Kingdom |
|---|---|
| Feb 18, 2026 | 1,987 |
| Feb 19, 2026 | 2,054 |
| Feb 20, 2026 | 1,936 |
| Feb 21, 2026 | 2,103 |
| Feb 22, 2026 | 1,978 |
| Feb 23, 2026 | 743 |
| Feb 24, 2026 | 828 |
| Feb 25, 2026 | 1,902 |
| Feb 26, 2026 | 2,041 |
| Feb 27, 2026 | 1,965 |
| Feb 28, 2026 | 2,088 |
| Mar 1, 2026 | 1,919 |
| Mar 2, 2026 | 767 |
| Mar 3, 2026 | 812 |
| Mar 4, 2026 | 1,994 |
| Mar 5, 2026 | 2,062 |
| Mar 6, 2026 | 1,927 |
| Mar 7, 2026 | 2,077 |
| Mar 8, 2026 | 1,948 |
| Mar 9, 2026 | 726 |
| Mar 10, 2026 | 851 |
| Mar 11, 2026 | 1,914 |
| Mar 12, 2026 | 1,950 |
| Mar 13, 2026 | 2,076 |
| Mar 14, 2026 | 2,090 |
| Mar 15, 2026 | 1,963 |
| Mar 16, 2026 | 705 |
| Mar 17, 2026 | 849 |
| Mar 18, 2026 | 1,923 |
| Mar 19, 2026 | 1,763 |
This view counts every session from UK users that includes at least one Add to Cart event, with one point per day. Read it left to right to see a weekly rhythm: sessions hold a weekday plateau around 2,000 and fall to weekend troughs near 700 to 800. The dashed segment at the right marks the most recent day, whose data is still incomplete, so treat that final point as provisional rather than a real drop.
The same chart with an added group segment by Carrier shows the segmented carriers as different colored lines.
| Date | Verizon | AT&T | T-Mobile | Sprint |
|---|---|---|---|---|
| Feb 18, 2026 | 781 | 592 | 328 | 263 |
| Feb 19, 2026 | 796 | 614 | 341 | 278 |
| Feb 20, 2026 | 773 | 576 | 319 | 257 |
| Feb 21, 2026 | 807 | 643 | 359 | 294 |
| Feb 22, 2026 | 762 | 601 | 333 | 268 |
| Feb 23, 2026 | 281 | 221 | 121 | 99 |
| Feb 24, 2026 | 318 | 258 | 128 | 103 |
| Feb 25, 2026 | 788 | 588 | 322 | 261 |
| Feb 26, 2026 | 774 | 631 | 347 | 283 |
| Feb 27, 2026 | 802 | 607 | 336 | 271 |
| Feb 28, 2026 | 759 | 573 | 314 | 254 |
| Mar 1, 2026 | 791 | 619 | 344 | 279 |
| Mar 2, 2026 | 274 | 214 | 119 | 97 |
| Mar 3, 2026 | 339 | 249 | 123 | 102 |
| Mar 4, 2026 | 783 | 603 | 331 | 266 |
| Mar 5, 2026 | 797 | 627 | 352 | 288 |
| Mar 6, 2026 | 768 | 581 | 318 | 258 |
| Mar 7, 2026 | 809 | 641 | 361 | 296 |
| Mar 8, 2026 | 751 | 596 | 329 | 264 |
| Mar 9, 2026 | 286 | 218 | 117 | 96 |
| Mar 10, 2026 | 347 | 263 | 126 | 104 |
| Mar 11, 2026 | 780 | 568 | 305 | 261 |
| Mar 12, 2026 | 767 | 581 | 334 | 268 |
| Mar 13, 2026 | 755 | 640 | 372 | 309 |
| Mar 14, 2026 | 807 | 650 | 338 | 295 |
| Mar 15, 2026 | 771 | 613 | 347 | 232 |
| Mar 16, 2026 | 274 | 214 | 119 | 98 |
| Mar 17, 2026 | 352 | 269 | 124 | 104 |
| Mar 18, 2026 | 771 | 571 | 332 | 249 |
| Mar 19, 2026 | 734 | 516 | 292 | 221 |
Each line tracks the same session count for one Carrier, which lets you compare segments at a glance. The carriers hold the same rank all month: Verizon sits highest, followed by AT&T, T-Mobile, then Sprint. Because every line dips together on weekends and none of them cross, the differences between carriers are proportional rather than a sign that one segment is overtaking another.
The data table
A table of session or event data appears below the chart. To specify which segments to see in the graph, click a segment name in the breakdown table. To download the table, click Export CSV.
Using the example above, Amplitude segments the breakdown table's results by Carrier.

Amplitude counts users as (none) if the segmented property values aren't available when the events trigger. Read more about (none) or unexpected values in this FAQ article.
The three ways Amplitude records sessions
Amplitude records sessions on either the server side or the client side. Client-side sessions can be either mobile or web.
- Server side: Use the HTTP API v2 to track sessions on the server side by including a value in the
session_idfield. Thesession_idvalue is the number of milliseconds since epoch, counting from the start of the session. - Client side (mobile): When you use Amplitude's mobile SDKs, events triggered within 5 minutes of each other count toward the current session by default. The time of the first event marks the session's start time, and the last event triggered marks the end time. For example, an 'Open App' event can mark the first event. Amplitude counts events sent within five minutes of each other toward the current session.
- Client side (web): When you use Amplitude's JavaScript SDK, events triggered within 30 minutes of each other count toward the current session by default. The time of the first event marks the session's start time, and the last event triggered marks the end time.
You can also define a session without instrumenting your events first, by setting a custom session property.
The User Sessions chart only displays data if you send a session ID with your events. Amplitude's SDKs handle this automatically, unless you flag an event as out-of-session (assigning the session ID a value of -1). If you're using Amplitude's HTTP API, you must explicitly send a session_id with your events.
How filtering works in the User Sessions chart
Filtering events for the User Sessions chart is a multi-step process. The order of those steps matters.
First, Amplitude filters for events that match the property filters. Amplitude then takes those events and groups them into sessions, which lets Amplitude calculate session length and count events performed each session.
Property filters apply before session filters. Amplitude filters on raw events first, then on the filtered events.
Amplitude only considers events with property filters when computing session length.
Was this helpful?