> For the complete documentation index, see [llms.txt](https://guide.data.gov.sg/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://guide.data.gov.sg/user-guide/for-data-owners/set-up-your-datasets-update-schedule-and-reminders.md).

# Set up your dataset’s update schedule and reminders

Use Data freshness to tell data consumers when to expect your next update and set email reminders for the people responsible for keeping the dataset updated.

{% hint style="info" %}
Set this up for each dataset you manage. A schedule records when you expect to update the data. You still need to upload the data yourself or maintain the pipeline that sends it.\
\
Data freshness is currently available to a limited number of agencies. We expect a full release at the start of 2027.
{% endhint %}

## Open the Data freshness settings

1. Sign in to the [data.gov.sg admin panel](https://admin.data.gov.sg).
2. Select your dataset from the dataset list.
3. Scroll to `Data freshness`.

{% hint style="info" %}
The `Public preview` beside the form shows how the update information will appear to data consumers.
{% endhint %}

<figure><img src="https://2014478147-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fksq6aNCGbng1fqooERfi%2Fuploads%2FqYXpBJh3qN4mFLeQkk9h%2Fimage.png?alt=media&amp;token=33bd3c45-10fd-421d-a627-22f063e67708" alt=""><figcaption></figcaption></figure>

## Set up an update schedule

1. Under “How is this dataset updated?”, select:
   * `Manual upload` if you upload each new version yourself.
   * `API pipeline` if a pipeline sends new data to data.gov.sg.
2. Under “Schedule type”, select `Recurring` if you update the dataset regularly. Select `No schedule or ad-hoc` if there is no fixed schedule.

<figure><img src="https://2014478147-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fksq6aNCGbng1fqooERfi%2Fuploads%2F19vGULO1th94f7AzuE71%2Fimage.png?alt=media&amp;token=d486cd76-7ce0-4cfc-985a-11036131d57d" alt=""><figcaption></figcaption></figure>

3. For a recurring schedule, use “Every” to enter how often you update the dataset.

<figure><img src="https://2014478147-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fksq6aNCGbng1fqooERfi%2Fuploads%2F1JWWaoutcyCrGeBE1h77%2Fimage.png?alt=media&amp;token=8d8a3646-8877-4884-b92c-0311e9fa4652" alt=""><figcaption></figcaption></figure>

| Your update frequency | What to enter                             |
| --------------------- | ----------------------------------------- |
| `Daily`               | Every 1 day                               |
| `Weekly`              | Every 1 week, then choose the day or days |
| `Monthly`             | Every 1 month, then choose “On day”       |
| `Quarterly`           | Every 3 months, then choose “On day”      |
| `Annually`            | Every 1 year                              |

API pipelines can also use minutes or hours.

{% hint style="info" %}
For monthly schedules, if you choose a day that does not exist in a particular month, the schedule uses the last day of that month. For example, a schedule set to the 31st falls on 30 April.
{% endhint %}

## Set the next expected update

1. Review “Next expected update” and choose the date you expect to provide the next update. Check any date filled in for you.
2. If you have a specific deadline, select `Set a specific time` and enter the time in Singapore time (SGT). Otherwise, the deadline defaults to 11:59PM SGT.
3. Choose a “Display setting” for the public dataset page.

<figure><img src="https://2014478147-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fksq6aNCGbng1fqooERfi%2Fuploads%2FFBz802rGoNeizQJ9udQ5%2Fimage.png?alt=media&amp;token=f387619d-84dc-4452-bcac-242c7dc921e3" alt=""><figcaption></figcaption></figure>

| Display setting     | What data consumers see                                                          |
| ------------------- | -------------------------------------------------------------------------------- |
| `Exact date`        | The expected update date                                                         |
| `Exact date & time` | The expected update date and time; available when you set a specific time        |
| `Month only`        | The expected month and year; available for schedules measured in months or years |
| `Hidden`            | **No next expected update date**; the recurring update frequency remains visible |

{% hint style="info" %}
For manual uploads, a next expected update date is required for schedules less frequent than daily. It is optional for daily uploads and API pipelines.
{% endhint %}

## Set up reminders for manual uploads

Reminders are available when you select `Manual upload` and a `Recurring` schedule.

1. In the “Reminder” section, turn on `Email notifications`.
2. Review “Admin notification emails”. Add the admins who should receive notifications, pressing Enter after each address. You can select `Same as contact emails` to use the dataset’s “Contact Emails” instead.
3. If someone else supplies the data, add their address under “Data provider emails”. This field is optional.
4. Under “Send first reminder”, choose how many hours, days or weeks before the deadline you want the first reminder.
5. To repeat reminders, select `Send follow-up reminders` and choose an interval.
6. Click `Publish changes` at the top of the page to save and apply the settings.

<figure><img src="https://2014478147-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fksq6aNCGbng1fqooERfi%2Fuploads%2FU1asCmCDonX3wrD71CdY%2Fimage.png?alt=media&amp;token=ddf92711-91e3-41e8-99d6-cbad1249b639" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
If the form warns that the reminder is longer than your update cycle, shorten the lead time or use the suggested value.
{% endhint %}

## Set up alerts for an API pipeline

For pipeline updates, use alerts to flag missing data after an expected update.

1. Select `API pipeline` and a `Recurring` schedule.
2. In the “Failure alert” section, turn on `Email alerts on missed updates`.
3. Review “Admin notification emails”. Add the people responsible for responding to missed updates, or select `Same as contact emails`.
4. Under “Data provider emails”, optionally add the pipeline owners, operations team or engineering team.
5. Under “Alert me if no data arrives”, choose how long to wait after the expected update before alerting. This grace period allows for small pipeline delays.
6. To repeat alerts, select `Send follow-up alerts` and choose an interval.
7. Click `Publish changes`.

<figure><img src="https://2014478147-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fksq6aNCGbng1fqooERfi%2Fuploads%2FH18H4wXzJT2d4kj1zMY0%2Fimage.png?alt=media&amp;token=db364c72-f09d-4491-9003-86ee4f18cf65" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
Datasets served as real-time APIs have different controls.
{% endhint %}

## Review and save your settings

Before clicking `Publish changes`, check:

* The schedule matches when your team can provide updates.
* The next expected date and any specific time are correct.
* The “Public preview” shows the detail you want to share with data consumers.
* Notification addresses and reminder or alert intervals are correct.

If you need to make edits to your schedule, return to your dataset’s “Data freshness” section, make the change, then click `Publish changes`.

## When you update your datasets before the schedule

If an “Update schedule” message appears after publishing, review the dates shown:

* Select `Keep [date]` if you still expect another update on the existing date.
* Select `Change to [date]` to use the proposed next update date.

### What data consumers see

Data consumers can see your recurring update frequency and, depending on your display setting, the next expected update date. The schedule describes your expected updates; “Last updated” records an actual data update for file-based datasets. Editing the schedule does not itself update the dataset’s contents.

For an explanation of the public information, see [Understanding a dataset’s update information](https://guide.data.gov.sg/user-guide/for-data-consumers/understanding-a-datasets-update-information).
