Webhook

Part of our topic guide on AI Skills for Business.

A webhook is an automated message one system sends to another the moment something happens: a payment clears, a form is submitted, an order ships. Instead of one system repeatedly asking "has anything changed?", the other system simply pushes the update straight away.

Push, not poll

The whole idea fits in one contrast. An ordinary API request is a pull: the receiving system has to keep actively asking for data ("polling"), and anything that happens between asks sits waiting to be noticed. A webhook flips the direction: the system where the event happened does the telling, sending the data out the moment it's ready. Nothing polls, nothing waits, nothing gets missed between checks.

One webhook, step by step

Follow a single event through and the mechanics are short:

1. The trigger. An event happens in the source system: a customer completes checkout, a support ticket is closed, a new row is added to a form.
2. The push. The source system sends an HTTP request (almost always a POST request) containing the event details to a URL the receiving system has registered in advance. This URL is often called an "endpoint" or "listener".
3. The payload. The request carries the relevant data, usually formatted as JSON, so the receiving system knows exactly what happened and to what record.
4. The response. The receiving system processes the data and sends back a simple confirmation (typically an HTTP 200 status) to say it arrived.

Where you'll meet them at work

Everywhere modern business tools talk to each other without a person copying data between them. Your payment provider tells your accounting software the instant an invoice is paid; your e-commerce platform tells your warehouse system the instant an order comes in. No one is manually checking, exporting or re-keying anything.

That's why the concept earns a place in a business glossary rather than a developer manual. So much of the daily grind in a data or operations role (chasing whether something has updated, reconciling two systems by hand) is exactly the kind of work a webhook removes. In our experience, the people who get real value from data and AI tools aren't the ones who know the most theory; they're the ones who understand how systems actually connect and where the manual steps still are. Most failed data and automation projects stall on people not knowing how to use what's already there, not on missing technology. For learners, webhooks are a good example of "the new basics": you don't need a computer science degree to understand one, but understanding how systems push and receive data is fast becoming as fundamental as reading a spreadsheet.