Automate a Workflow on a Schedule

~25 minutes · a scheduled run, a run history, a handled failure

You already have a workflow — say the sales-cleanup pipeline from Load, Transform, and Publish Data. Right now it runs when you click Run. This tutorial hands that job to PlaidCloud: the workflow runs on a calendar, on its own, and you learn to watch it, prove it ran, and recover it when a run fails.

Three things turn a manual pipeline into an automated one, and this tutorial does each in turn:

A schedule — a rule that fires the workflow at the times you choose.
A record you can trust — run history and logs that tell you what happened while you weren’t watching.
A failure plan — because an unattended job eventually hits bad data, and it should not fail silently.

Once a schedule owns the run, the history and log are how you know it worked — and Resume is how you fix it when it didn’t.

Prerequisites

A PlaidCloud workspace with a workflow that already runs correctly when you run it by hand. If you don’t have one, build it first with Load, Transform, and Publish Data.
Access to Analyze → Tools → Event Scheduler in that workspace.

Schedule a workflow that already passes. Automation repeats whatever the workflow does — including its mistakes, now unattended. Run it manually to a clean finish once before you put it on a schedule.

Step 1: Confirm It Runs by Hand

Open your workflow and run it once, the manual way, so you have a baseline to compare against.

Open your project and go to the Workflows hierarchy.
Click the run icon on the workflow, or open the workflow and choose Run All from the Actions menu. (The Toggle Start/Stop button at the top of the workflow table does the same, and stops a run in progress.)
Watch the step status icons as the run moves through. Each step shows where it stands and how long it has been running — hover an icon for the message behind it.

The statuses you’ll see, and what each means:

Status	What it means
Waiting	Queued, not started yet.
Running	Working; its duration counts up.
Completed	Finished successfully.
Warning	Finished, but flagged something worth reading in the log.
Error	The step failed — see Managing Step Errors.
Skipped	Passed over — disabled, or its run conditions weren’t met.
Continued	Failed but set to continue on error, so the run carried on.

A clean run ends with every step Completed. That’s your green baseline. Full status details are in Run a Workflow.

Step 2: Create the Schedule

Now hand that run to the Event Scheduler.

Open Analyze, select Tools, and click Event Scheduler. The Events Table lists every event already configured for the workspace.
Click Add Scheduled Event.
Complete the required fields the form marks, pointing the event at the workflow you just ran (and its project).
Open the Event Schedule tab and pick a Schedule type:
- Use Classic Schedule — choose the specific months, days, hours, and minutes to run. Start with something you’ll see happen soon, like a few minutes from now.
- Use Workday Schedule — choose a financial workday instead of a clock time. Workday schedules can also honor holiday calendars, so a run lands on the right business day.
Click Create.

The Events Table now lists your event, and its description column shows whether it’s scheduled by month, by day/hour/minute, or by workday.

Times are UTC unless you say otherwise. New events default to the UTC timezone. If “9 a.m.” needs to mean 9 a.m. where you are, set the Timezone field on the event — otherwise a scheduled run can land hours off from what you expected.

Cap a test schedule so it can’t run forever. The Limit Running section runs an event only for a set time period and a set number of times. On a schedule you’re just trying out, that’s how you avoid a job that quietly fires every few minutes for weeks.

Step 3: Preview When It Will Run

Before you trust the schedule, look at what it will actually do.

Open the Upcoming Runs Calendar. It’s a read-only preview — it expands the same schedules that drive real execution, so what you see is what will run, and looking never triggers anything.
Switch between Month, Week, and Agenda views; use Previous, Next, and Today to move through time, and Refresh to re-pull the latest.
Confirm your event appears when you expect it. Each run is drawn as a bar sized to the workflow’s typical duration.

Side-by-side bars mean an overlap. If two runs’ bars sit next to each other in the same window, those runs collide — stagger their schedules if they compete for the same data or resources. Event-driven sensors don’t appear here; they have no fixed future time. See Upcoming Runs Calendar.

If you’d rather not wait for the clock, select the event in the Events Table and click Run Selected Events to fire it immediately — a good way to prove the scheduled path works end to end right now.

Step 4: Watch the Run History and Log

An automated run happens while you’re elsewhere, so the record it leaves is what you actually rely on.

Run history lives with the workflow. Each execution is its own entry. A resumed run is recorded as a new run containing only the steps it re-ran, and the original keeps its own entry — so history reads as a truthful sequence of what happened, not a single line that gets overwritten.
The workflow log captures events as they happen — steps running, warnings raised. Open it from the Project area under the Log tab; it also rolls up into the broader project log so you can watch several workflows at once. The viewer sorts and filters, and you can open any entry for detail. See Viewing Workflow Log.
The workflow report documents the workflow’s current state — its steps, functions, and variables. It’s generated on demand from the Report icon in the Workflows hierarchy, which is handy when someone asks what an automated job actually does. See View Workflow Report.

After a scheduled or on-demand run, open the log and confirm the run you expected is there and finished as you intended.

Step 5: Handle a Failure

An unattended job will eventually meet bad data. Decide now what should happen when it does — PlaidCloud gives you three responses, and the right one depends on the step.

Choice	What happens	Use it when
Stop the workflow	The run halts; an error indicator shows on both the step and the workflow.	The default — most steps should not proceed past a real failure.
Continue on error	The step is marked with an error, but the run treats it as done and carries on.	A step that can harmlessly fail on missing data.
Remediation workflow	The failing run stops and a second workflow starts instead.	You want a failure to notify someone or attempt an automatic fix.

Two settings shape this, both on the step’s edit form (the pencil icon in the workflow table):

Retry — a failed step can retry a set number of times, with a delay between attempts from seconds to hours. This alone rescues most failures that come from a remote system being briefly unavailable. See Managing Step Errors.
Continue On Error — check this box and the step’s failure won’t stop the workflow. Steps set this way carry a special indicator in the workflow table. See Continue on Error.

Continue-on-error hides the failure from your remediation plan. A step marked to continue never stops the workflow, so it never triggers a remediation workflow either — only a failure that would otherwise halt the run does that. Reserve continue-on-error for failures you’re genuinely fine ignoring.

When a scheduled run does fail, you rarely need to start over. Resume picks a stopped or errored run back up from where it left off — it re-runs the steps that didn’t finish successfully and leaves completed ones alone. Resume is available whenever a workflow’s last run ended paused, stopped, or in error.

Step 6: Pause It When You Need To

Automation you can’t turn off is a liability. To take a schedule offline without deleting it:

Open Analyze → Tools → Event Scheduler and click the edit icon on the event.
Uncheck the Active checkbox and click Update.

The event stops firing on its schedule until you re-activate it — the same edit path, with Active checked again. To remove it for good, use the delete icon instead. An active event shows the Active icon in the Events Table, so a glance tells you what’s live.

You’re set. You now have a scheduled event, a run history to watch, and a way to handle the first failure — and you can pause the whole thing without deleting it.

Where to Go Next

Event Scheduler — every field on a scheduled event, including workday and holiday-calendar scheduling.
Scheduled Workflows — run windows, ordering, and conditional triggers across a workspace.
Upcoming Runs Calendar — spot colliding run windows before they compete.
Run a Workflow — pause, stop, and resume in full, plus every step status.
Managing Step Errors and Continue on Error — the failure-handling settings in depth.
Workflows guide — conditions, loops, and variables for more advanced automation.
Load, Transform, and Publish Data — the tutorial that builds the workflow this one automates.