Economic Calendar Scraper: Actual, Consensus, Forecast & Previous

Scrape the economic calendar of 140+ countries into clean rows. You get every macro release, including CPI, inflation, non-farm payrolls, unemployment, GDP, PMIs, retail sales, central-bank rate decisions, bond auctions, holidays and elections. Each row has the release time in UTC, country, currency, reference period, importance (1–3), actual, consensus, forecast and previous values (as shown and as numbers), the revised previous value, the unit and the surprise (actual minus consensus). Pick a date preset or your own dates, filter by country, currency, region, importance and keywords, and schedule it with Only new or updated events to catch each release as it comes out. No login, no API key.

What you get

When	Release date and time in UTC (2026-10-27T12:30:00Z), all-day flag for holidays and tentative releases
Where	Country, ISO code (US, EA, GB, JP …), currency (USD, EUR, GBP, JPY …)
What	Event name ("Durable Goods Orders MoM"), indicator category, reference period ("SEP", "Q3", "OCT/24"), ticker, link to the indicator page
Importance	1 (low), 2 (medium) or 3 (high, market-moving)
Values	Actual, consensus, forecast and previous as published ("1.8%", "$-52.8B", "194K") and as numbers, unit and currency sign, the previous value before its revision
Surprise	Actual minus consensus, when both are known in the same unit
Status	Released (has an actual value) or upcoming; in monitor mode, new or updated plus which fields changed

How to use it

1. Enter Countries: ISO codes (US, EU, GB, JP, CA, DE, CN), names (United States, Euro Area), currencies (USD, EUR; EUR is the euro area plus every euro country) or regions (G20, G7, Europe, Asia, America, Africa). Leave it empty for every country.
2. Pick Dates: today, tomorrow, yesterday, this week, next week, last 7 days, this month, next month or last 30 days. Or fill in From / Until (e.g. 2026-11-02, or -7 days, +14 days).
3. Optionally set Minimum importance, Event keywords (CPI, non farm payrolls, interest rate decision, PMI), Released / upcoming and Include holidays.
4. Set Max events and run.

The calendar covers the last 30 days through the end of next month. Days outside that window are skipped, and the status message says so.

Input examples

This week's US events (the example input):

{
  "countries": ["US"],
  "dateRange": "thisWeek",
  "maxItems": 20
}

Next week's high-importance releases for the major currencies:

{
  "countries": ["USD", "EUR", "GBP", "JPY", "CAD", "AUD", "CHF", "NZD"],
  "dateRange": "nextWeek",
  "minImportance": "high",
  "includeHolidays": false,
  "maxItems": 0
}

Every inflation and jobs release in the G20 over custom dates:

{
  "countries": ["G20"],
  "dateFrom": "2026-11-01",
  "dateTo": "2026-11-30",
  "keywords": ["inflation rate", "CPI", "non farm payrolls", "unemployment rate"],
  "maxItems": 0
}

A release monitor: schedule this every 5–15 minutes and each run saves only the events that were just released or changed:

{
  "countries": ["US", "EU", "GB"],
  "dateRange": "today",
  "minImportance": "medium",
  "onlyChanges": true,
  "maxItems": 0
}

Output example

{
  "eventId": "398125",
  "date": "2026-10-27",
  "time": "12:30",
  "datetimeUtc": "2026-10-27T12:30:00Z",
  "allDay": false,
  "country": "United States",
  "countryCode": "US",
  "currency": "USD",
  "event": "Durable Goods Orders MoM",
  "category": "durable goods orders",
  "period": "SEP",
  "importance": 3,
  "importanceLabel": "high",
  "actual": "1.8%",
  "consensus": "1.5%",
  "forecast": "1.3%",
  "previous": "0.1%",
  "previousRevisedFrom": "0%",
  "actualValue": 1.8,
  "consensusValue": 1.5,
  "forecastValue": 1.3,
  "previousValue": 0.1,
  "unit": "%",
  "valuePrefix": null,
  "surprise": 0.3,
  "released": true,
  "ticker": "UNITEDSTADURGOOORD",
  "url": "https://tradingeconomics.com/united-states/durable-goods-orders",
  "source": "Trading Economics",
  "changeType": null,
  "changedFields": [],
  "scrapedAt": "2026-10-29T18:53:13.069Z"
}

Every row has the same columns. A value the calendar doesn't show is null. For example, actual stays null until the release.

Pricing

Pay per event: $1.50 per 1,000 events saved, plus $0.002 per run. Wide searches cost a little extra. The calendar shows at most 1,000 events per page, so a range with more events than that (every country over a whole month), or custom dates that span two of the calendar's ranges, needs more pages. Each page after the first costs $0.001. A typical run reads one page and never pays that. Events skipped by your filters or by Only new or updated events are never charged.

Examples: this week's US calendar (about 90 events) ≈ $0.13. Every country for a week (about 570 events) ≈ $0.86. A 5-minute monitor that finds nothing new costs $0.002 per run.

Use cases

• FX and macro trading: the week ahead with consensus and forecast, and the surprise the moment a number is released.
• Trading bots and alerts: a scheduled monitor run that saves only new and updated events, ready for a webhook, Slack or Telegram.
• Newsletters and dashboards: next week's market-moving releases by country, ready to paste into a newsletter or load into a dashboard.
• Research: actual versus consensus surprises across countries for the last 30 days.

Tips

• Use Minimum importance: High only for the releases that move markets (CPI, payrolls, rate decisions, GDP).
• Only new or updated events remembers what earlier runs with the same input saved. An event is saved again (with changeType: "updated" and changedFields) when its actual, consensus, forecast, previous value or time changes.
• Times are always UTC. Convert with datetimeUtc in your own tool.
• surprise is only filled when actual and consensus use the same unit.

FAQ

Do I need an account, an API key or a login? No. The scraper reads the same public calendar any visitor sees.

How far back and ahead does it go? From 30 days ago to the end of next month, which is what the calendar shows without an account. Older history is not available here.

How quickly does a new number show up? As soon as the calendar publishes it. For a release monitor, schedule a run every few minutes with Only new or updated events. Each run costs $0.002 plus the events it saves.

Why did a run say some events may be missing? The calendar sometimes answers with a different selection of countries. The scraper checks every answer and asks again. If the calendar keeps sending the wrong selection, the run keeps only the matching events and says so in its status message; if it happens on two pages of one run, the run stops early rather than spend your time on more of them. Running it again a few minutes later usually completes them.

Which proxy should I use? Keep the default (Apify Proxy, datacenter). It comes with every Apify plan, including the free one.

Is it legal to scrape Trading Economics? This Actor only collects publicly available data: the economic calendar that anyone can see on Trading Economics without an account (release times, countries, indicator names and the published values). It contains no personal data. Collecting publicly available data is generally legal, but you're responsible for how you use it and for following Trading Economics' terms. If you're unsure, check with a lawyer. More on this: Is web scraping legal?

Does it access any private data? No. Everything comes from the public calendar page that is shown without logging in. The scraper never touches accounts, never uses paid data feeds and follows the site's robots.txt.

This Actor is not affiliated with, endorsed by or sponsored by Trading Economics.