What do the visits without duration correspond to?

A visit without duration in Piano Analytics is typically a visit with only one event.

As a reminder, a visit begins at the first event recorded, and the time spent is based on the difference in time between events. If only one event is recorded during a visit, Piano Analytics has no second timestamp to compare against, so there is no way to calculate how long the visitor stayed on the site without generating another event.

How visit duration is calculated

A visit is considered to have a measurable duration only when at least two distinct events are recorded in that visit.
Duration is derived from the difference between event timestamps across the visit (from the first measured event to the last), expressed in seconds.
If no additional event occurs, the visit duration cannot be computed, and the visit is treated as without duration in standard reporting.

When you may see “no duration” even with multiple events

In some cases, a visit can contain at least two events but still appear as having no duration in reports due to rounding:

Visit duration is rounded to the nearest whole second.
If the time between the first and last event rounds to 0 seconds (for example, two events occurring within the same civil second), the visit may be treated as having no duration in standard reports.
If events span different seconds, even if the actual time difference is less than one second, rounding can result in a duration of at least 1 second.

Visit boundaries and inactivity timeout

A visit starts as soon as a visitor generates an on-site event. By default, a visit ends after 30 minutes of inactivity (no events). If a new event occurs after that inactivity period, it is counted as a new visit.

Related metrics and common reporting differences

The Visit Duration property represents the visit duration in seconds, calculated from the first measured event to the last.
Metrics such as Time Spent per Visit are derived from time-spent calculations and may apply additional rules (for example, excluding visits whose duration rounds to 0 seconds), which can lead to differences versus raw visit duration distributions.

If you are comparing results across tools (for example, Data Query vs. Data Flow), ensure you group data by visit identifier and account for the requirements above (at least two events and a duration of at least 1 second after rounding when applicable).