> For the complete documentation index, see [llms.txt](https://docs.kernel.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.kernel.ai/platform/data-setup/scheduled-inbound-flows.md).

# Scheduled inbound flows

Configure recurring Salesforce imports with saved mappings, a CRM filter, and a schedule.

Import selected Salesforce records into Kernel on a recurring schedule, then write results using your saved output mapping.

A scheduled inbound flow combines a source mapping, a record filter, an output mapping, and a schedule. Kernel runs the saved query each time the schedule runs and processes eligible records using its built-in inbound processing.

This is separate from [Automatically fetch CRM changes](https://docs.kernel.ai/platform/crm.md#automatically-fetch-crm-changes), which updates Kernel’s copy of CRM data to reassess risk, flag identity drift and update Kernel’s recommendations. Scheduled inbound flows select and process records using their own saved filters and schedules.

[Create accounts](https://docs.kernel.ai/platform/actioning/create.md) is a separate action for adding missing accounts to your write source.

## Before you begin

Complete your Salesforce connection setup and prepare:

* A saved import mapping for **Account** or **Kernel Account**, covering the fields Kernel should read.
* A saved output mapping for the results Kernel should write.
* A SOQL filter selecting the records you want to process.

Configure mapping fields under **Connections > Salesforce > Data flows** before creating the inbound flow. [Connections and sync](https://docs.kernel.ai/platform/crm.md#data-flows) explains the import and output editors.

The pull mapping determines the Salesforce object and fields Kernel reads. The output mapping determines where results go. **Record Sync** mappings are not available as inbound-flow output destinations.

## Create a scheduled flow

Open [Connections > Salesforce > Inbound flows](https://app.kernel.ai/settings/connection/salesforce?tab=inbound-flows), then select **New inbound flow**.

New flows use **Scheduled inbound**. Complete the configuration:

1. **Name:** Enter a name identifying the flow's purpose.
2. **Enabled:** Choose whether the flow should run. This is on by default.
3. **Pull source:** Select the saved Account or Kernel Account import mapping.
4. **Push destination:** Select the saved output mapping for the results.
5. **Schedule:** Choose **Minute**, **5 minutes**, **Hourly**, or **Daily**. The default is Daily. **Off** leaves the flow without an active schedule.
6. **CRM where clause:** Enter the SOQL filter for the selected source object. It is required and must begin with `WHERE`.
7. Select **Save**.

The name, pull source, push destination, and CRM where clause are required. New scheduled flows use built-in processing; a separately linked enrichment workflow is not required.

## Choose a filter for repeated runs

Write the filter against the Salesforce object selected by the pull mapping. Use that object's Salesforce API field names.

For a first check, select one known record:

```sql
WHERE Id = '<Salesforce record ID>'
```

Replace the placeholder with a record ID from the selected source object.

For example, an Account flow could select records created in the past seven days:

```sql
WHERE CreatedDate = LAST_N_DAYS:7
```

The flow runs the saved query again on each scheduled run. A date condition is evaluated again at that time; it does not become a cursor automatically advancing past previously selected records. Processing eligibility, cooldowns, and attempt limits also affect which returned records run.

Choose a filter that remains appropriate when repeated. Check the selected records and output mapping before enabling a wider selection.

## Pause or resume a flow

Use **Enabled** in the list or configuration sheet to stop or resume scheduled runs. When you turn it off, runs already in progress finish; new scheduled runs do not start.

## Check the flow's status

For scheduled flows:

| Status             | Meaning                                                                                                            |
| ------------------ | ------------------------------------------------------------------------------------------------------------------ |
| **Live**           | The flow is enabled, has an active schedule, and has an output mapping or a linked legacy workflow.                |
| **Off**            | An active schedule is configured, but Enabled is off.                                                              |
| **Not configured** | The flow type or schedule is missing, or Schedule is set to Off.                                                   |
| **Awaiting**       | An active schedule is configured and Enabled is on, but no output mapping or legacy enrichment workflow is linked. |

New scheduled flows require an output mapping and use built-in processing. Workflow-linking guidance applies to existing legacy configurations.

**Live** describes configuration readiness. Check the selected destination to confirm that the expected mapped results have arrived.

## Edit a flow and review changes

Select an existing flow to open **Configuration**. Review its name, enabled state, saved mappings, schedule, and filter, then select **Save** after making changes.

Use **History** to review configuration changes, including who changed a setting and its previous and new values. This tab records configuration changes rather than every processing run.
