Plans and permissions:
- Plan: Enterprise plan
- Create an API key to load data yourself: Admin role, or a custom role with Assembled API keys set to Create, edit, and delete API keys
- The plan applies to the Snowflake data share. Loading data from the Assembled API works on any plan.
Use the Snowflake data share to build your own reports and dashboards on Assembled’s workforce metrics, in your own Snowflake account. Assembled shares a set of secure views with your account, so you can query your metrics next to your other data without exporting anything.
To get the data share, contact your Assembled representative or our team.
How the Snowflake data share works
The share contains three views:
| View | What it contains |
|---|---|
ASSEMBLED_METRIC_MAPPING |
Every metric in the share, with its title and description. For metrics calculated from other metrics, it lists the fields used in the calculation. |
ASSEMBLED_METRICS |
Values for metrics computed directly from your data |
ASSEMBLED_METRICS_DERIVED |
Values for metrics calculated from other metrics, such as adherence percentage |
Data in these views is broken down by hourly interval, team member, queue, channel, and metric.
What the data share doesn’t include
- Live data. Data in the share can be several hours behind, so it doesn’t suit intraday decisions.
- Anything that isn’t a metric in
ASSEMBLED_METRIC_MAPPING, such as scheduled events, forecasts, or activities. For those, including upcoming shifts, use the Assembled API.
Load Assembled data into Snowflake yourself
You don’t need the data share to get Assembled data into Snowflake. Many teams call the Assembled API from their own script or ETL tool and load the results into a Snowflake table they manage. This works on any plan, and you control which data you load and when.
- To get scheduled events, adherence totals, or working hours, see Scheduled hours in the Assembled API.
- Reports are built in the background: the first call returns only a report ID. If your table shows report IDs instead of metrics, have your pipeline poll GET /v0/reports/:reportID until the report is complete, then load it.
- For every endpoint and report type, see the Reports API reference.