You can automate your Stripe Sigma queries by scheduling them to run on a daily, weekly, or monthly basis. Results for each scheduled query are sent via email to specified team members or as webhook events.

Scheduling a query

With a query loaded into the editor, click Schedule. We recommend you uniquely name all your scheduled queries to avoid confusion. If your query doesn’t already have a name (or you wish to modify it), you can update it during the scheduling process.

Schedule

Each scheduled query can be run on a daily, weekly, or monthly basis. Queries run as soon as the data for that period is available.

Subscribers

Creators of scheduled queries are added as subscribers to email notifications by default. To notify other team members as well, enter their email addresses. Results sent via email include the name and date of the scheduled query, and a link to download the results in CSV format. To preview what the email looks like, click Preview email.

You or your team members can stop receiving notifications at any time by clicking the Unsubscribe link in the email. You can also edit the scheduled query in the Dashboard and add or remove subscribers.

Timeline

Based upon your chosen schedule, the timeline displays the date your query will next run, and the processing date of the data it will use (additional time is required to make your account data available to query).

Managing scheduled queries

Upcoming scheduled queries are displayed under Scheduled within the Queries tab. Schedules are grouped based on whether they were created by you or other members of your team.

To edit a scheduled query, select it and click Edit schedule. To delete it, click ••• and select Delete.

Receiving results as webhooks

If you make use of webhooks, you can receive notifications for scheduled queries as webhook events. Stripe sends the sigma.scheduled_query_run.created event each time a scheduled query is run. See below for a sample event.

The data.object.file.url subfield of the webhook payload contains the URL from which you can access the results file using your live secret API key. For example, if your server received the webhook below, it could download the results using this curl command:

curl https://files.stripe.com/v1/files/file_KCpwjYx2/contents -u sk_live_XXXXXXX:

For more on how to integrate webhooks, see our webhook documentation and especially the section on testing.

// Sample payload of a sigma.scheduled_query_run.created webhook

{
  "object": "event",
  "pending_webhooks": 2,
  "created": 1504794194,
  "type": "scheduled_query_run.created",
  "livemode": true,
  "request": null,
  "data": {
    "object": {
      "id": "sqr_B4rIp6TsiuLBdblsY9JX",
      "object": "scheduled_query_run",
      "status": "completed",
      "scheduled_query": "scq_wt9MjsH5cX2Atfj4rUmD",
      "data_load_time": 1504656000,
      "file": {
        "id": "file_KCpwjYx2",
        "object": "file",
        "url": "https://files.stripe.com/v1/files/file_KCpwjYx2/contents",
        "created": 1507841188,
        "purpose": "sigma_scheduled_query",
        "size": 53075,
        "type": "csv"
      },
      "title": "Scheduled Query Example",
      "sql": "SELECT count(*) FROM charges WHERE created >= date('2017-01-01')",
      "created": 1504794194,
      "completed": 1504794133,
      "result_available_until": 1505398933,
      "error": null,
      "livemode": true
    }
  }
}

Referencing data freshness

Additional time is required for data to be made available to query. You can use data_load_time as a value in your queries that represents when data was most recently processed on your account. This is useful for dynamically setting a date range in your scheduled queries.

For example, if the current value for data_load_time is 2019-06-29, the following query returns a list of balance transactions created one month before this date.

  select
  id,
  amount,
  fee,
  currency
from balance_transactions -- this table is the canonical record of changes to your Stripe balance
where
  created < data_load_time and
  created >= data_load_time - interval '1' month
order by created desc
limit 10

If this query is scheduled to recur weekly, the value for data_load_time on the next occurence would be 2019-07-06.