Skip to content
Tutorials Center
My learning

Topics

Lesson 8 of 38 ยท Flows and Triggers

Scheduled Flows: The Recurrence Trigger and Time Zones

Create scheduled cloud flows with the Recurrence trigger: intervals, frequencies, start times, time zones, weekdays and hours, delay actions and common timing pitfalls.

A scheduled cloud flow runs on a clock. It suits reports, reminders, clean-up jobs and data synchronization: any work that should happen whether or not something else triggered it.

The Recurrence trigger

Create a Scheduled cloud flow and set Starting, Repeat every and the unit, or add the Recurrence trigger yourself. Its fields:

Field Meaning
Interval The number, such as 1 or 15
Frequency Second, Minute, Hour, Day, Week or Month
Time zone The zone the schedule uses
Start time An ISO time such as 2025-03-03T07:00:00Z
On these days Weekdays, when frequency is Week
At these hours / minutes Exact hours and minutes, for Day and Week

Examples:

  • Every weekday at 7:30: frequency Week, interval 1, days Monday to Friday, hours 7, minutes 30.
  • Every 15 minutes: frequency Minute, interval 15.
  • First of each month: frequency Month, interval 1, with On these days set in the advanced parameters (month days).

Minimum frequency depends on your plan; free and low plans may limit how often a flow can run.

Time zones and daylight saving

Without a time zone the recurrence uses UTC, so a flow set to 7:00 will drift against local time when daylight saving changes. Pick the Time zone in the Advanced parameters so a 7:00 run stays at 7:00 local time.

Calculations inside the flow still use UTC by default: utcNow() returns UTC. Convert when you need local time (see Text, dates and time zones).

Start time and what "starting" means

The Start time anchors the schedule. A flow set to start tomorrow at 9:00 with an interval of 1 day runs at 9:00 every day. If you save a flow with a start time in the past, the schedule counts forward from that anchor, so a 4-hour interval anchored at 8:00 runs at 12:00, 16:00 and so on regardless of when you saved.

Runs can start a little after the scheduled time under heavy load. Do not rely on second-exact timing.

Other ways to wait

  • Delay pauses for an interval (for example 5 minutes).
  • Delay until pauses until a specific timestamp, which is useful for "send the reminder a day before the event".
  • Delay actions count against the flow's total run duration limit, which is 30 days for a run. Use schedules rather than very long delays for long waits.

Simulating a schedule

This stand-in uses Python's standard library to list the next run times for "weekdays at 07:30". It shows the logic a recurrence encodes. It is not the Power Automate scheduler and ignores time zones for simplicity.

Python Stand-in: next five weekday 07:30 runs

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Notice that the first run is the next day because 7:30 on March 6 has already passed.

Interval schedules

An interval schedule counts from the anchor, not from the end of the previous run. This stand-in lists runs for "every 4 hours from 08:00".

Python Stand-in: every 4 hours from an anchor

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Typical scheduled jobs

  • Daily digest: query a list for items due today, build a summary, send an email (see Arrays, objects and JSON).
  • Housekeeping: delete files older than 30 days.
  • Sync: copy new rows from one system to another and remember a high-water mark (the last id or time processed).
  • Health check: call an API and alert when it does not answer.

When a scheduled flow can take longer than its interval, add a safeguard. If two runs overlap they may process the same data twice. Turn on concurrency control with a degree of 1 in the trigger settings, so a new run waits for the previous one.

Common mistakes

  • Forgetting the time zone and being an hour out after daylight saving changes.
  • Scheduling a flow every minute that only needs to run daily, wasting your daily request quota.
  • Not handling days without data: a report flow that fails when the list is empty.
  • Assuming a skipped run is caught up later. Missed runs (for example because the flow was off) are not replayed.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Which Recurrence field keeps a 7:00 schedule at 7:00 local time through daylight saving changes?

    Setting the time zone makes the schedule follow local time instead of UTC.
  2. 2. What does utcNow() return?

    utcNow() always returns UTC; convert it when you need a local time.
  3. 3. How do you stop two overlapping runs of a scheduled flow?

    A degree of 1 makes a new run wait until the previous one is complete.
  4. 4. What happens to runs missed while a flow was turned off?

    Scheduled runs that did not happen are not caught up later.

Summary

  • The Recurrence trigger takes interval, frequency, time zone, start time and optionally weekdays, hours and minutes.
  • Always set the time zone for human-facing schedules.
  • Use Delay and Delay until sparingly; prefer a schedule for long waits.
  • Guard against overlapping runs and empty data.

Next: Trigger conditions and filtering