Official statistics from 29 agencies, now inside Datastam
Datastam now searches 2,190 series from 29 official statistical agencies, including StatCan, FRED, Eurostat and the ECB, and analyzes any of them in one click.
Datastam started as a place to analyze your own spreadsheet. Most questions worth asking, though, need a public number next to your private one: your sales against the consumer price index, your hiring against the unemployment rate, your export mix against trade data.
Those numbers are now built in. Official statistics lets you search series from 29 national and international statistical agencies and analyze any of them in one click, without leaving your workspace.
What you can do
Ask directly. Type a question such as "How has Canadian inflation compared to the US since 2020?" and get a computed answer, a chart and a citation to the official table. No dataset to download first.
Search and import. Search the catalog by indicator ("housing prices", "unemployment", "GDP") or browse by source. Click "Start Analyzing" and the series lands in a new conversation with the same automatic analyses a dropped CSV gets: overview, trends, distributions, anomalies. From there it is a normal Datastam chat: ask follow-ups, build charts, export a PDF, publish a story.
Add to an existing chat. Inside any conversation, the attach menu has an "Official statistics" option. Pull in the CPI next to your own revenue file and ask how they move together.
Browsing and searching are open to everyone. You will be asked to sign in when you import or ask, and imports count toward your plan's daily uploads.
The 29 sources
The connector is powered by Starwell, an independent Toronto company that harmonizes official statistics into one API and checks the values it serves against the publishing agency. As of today it exposes these sources:
| Region | Sources |
|---|---|
| Canada | Statistics Canada |
| United States | FRED (Federal Reserve Bank of St. Louis), Bureau of Labor Statistics, Bureau of Economic Analysis, Treasury Fiscal Data, SEC EDGAR, Bureau of Transportation Statistics, CDC (NCHS) |
| United Kingdom | Office for National Statistics |
| Europe | European Central Bank, Eurostat, INSEE (France), Statistics Netherlands, INE (Spain), Statistics Austria, Statistics Poland, Statistics Sweden, Statistics Norway, Statistics Denmark, Statistics Finland, Statistics Iceland, Statistics Estonia, CSO Ireland, Czech National Bank |
| Asia-Pacific | Australian Bureau of Statistics, Department of Statistics Malaysia |
| International | World Bank, OECD, ILOSTAT |
Together they cover 260 curated datasets and 2,190 series, and the list grows as Starwell adds connectors. Statistics Canada is the deepest source today, followed by the World Bank and the European Central Bank.
Series count per source served through the connector. Source: Starwell public API, /sources, retrieved September 15, 2026. Counts are the curated subset each agency exposes through Starwell, not the agency's full catalog.
Where the numbers come from, and how they are credited
Datastam calls Starwell's public REST API the way any customer would. Each series carries the URL of the official table it was fetched from, when it was retrieved, and a verification status; results that pass Starwell's check against the live source show a "verified" badge in the picker.
When you import a series, Datastam stores a copy in your workspace as a CSV and records the publishing agency's licence and attribution as the first message in the conversation. The source chip on every chart names the agency itself, such as "Statistics Canada" or "Federal Reserve Economic Data (FRED)". Series whose licence does not permit redistribution are not importable.
Two things worth knowing. An import is a snapshot: if the agency revises the series later, re-import to pick up the change. And the catalog is a curated subset of each agency's publications; the page says so next to every source count.
What about canada.datastam.ai?
Nothing changes there. The Canada Data Explorer talks to Statistics Canada directly and covers all 8,270 of its tables. Official statistics is for the main Datastam workspace, where you bring your own data and want the official series beside it.
Try it at datastam.ai/official-data.