An API is a door an app opens for software: you send a request and get clean data back, usually as JSON, with a key. Web scraping reads a web page the way a person would and pulls text out of the HTML. Use the official API whenever there is one, an RSS feed when a site offers it, a pay-per-call data API for things like search results, and scraping only as a last resort, because scrapers break when the page changes and are often blocked. The free lesson APIs vs Scraping vs RSS shows all four side by side, and it opens a free 10-module course, Get Data From the Web Into Your Automations, where you practice each one in Coefficient and n8n.
This page is for office, marketing and operations people who copy prices, news, CRM numbers or Reddit posts into a report by hand, and for anyone who has built a first automation and now wants outside data in it. No coding is needed. Every example uses made-up businesses and data.
Why does the data source matter more than the tool?
Every Monday, someone at a small firm copies numbers from websites and apps into a sheet. It takes an hour or two, it is the first job skipped in a busy week, and a copy slip becomes a wrong number in a report that goes to a client.
The automation tool is rarely the hard part. n8n, Zapier and Make can all move data into a sheet. What breaks workflows is the source: a page redesign that kills a scraper, a site that blocks you, a free feed that is switched off, a key in the wrong place, a list that quietly stops at page one. Pick the source well, learn a few API habits, and the workflow keeps running for years with one small change now and then.
What are the four ways to get web data into an automation?
| Way in | What it is | Best for | Watch out |
|---|---|---|---|
| Official API | The app's own door for software | Your own app data: CRM, sheets, email tools | Needs a key; read the doc |
| RSS feed | A free list of a site's newest posts | News, blogs, podcasts | Sites can switch feeds off |
| Third-party data API | A company that collects data and sells it per call | Search results, shopping prices, Reddit posts | Costs per call; multiply first |
| Web scraping | Software reads the page like a person | Sites with no other way in | Breaks, gets blocked, may break terms |
In the course's first real module, APIs vs Scraping vs RSS, you pick the way in for three jobs at a made-up agency. Industry news from a site with a feed: RSS. Open deals from the agency's own CRM: the official API, or a connector built on it. Competitor prices from Google Shopping: a search-results API that charges per call, because Google blocks scrapers and offers no feed for it.
What is an HTTP request?
An HTTP request is the message your automation sends to an API. It always has the same five parts:
- A method. GET reads data. POST sends data in or runs a search.
- A URL, the endpoint. One address for one kind of data, such as /v1/deals.
- Query parameters. Filters after the question mark, such as ?status=open&limit=50.
- Headers. Who is asking. The key usually goes here as a bearer token.
- The response. A status code (200 means it worked) and the data as JSON.
n8n's HTTP Request node, Zapier's webhook step and Make's HTTP module are forms for these five parts. In Call Any API with the HTTP Request Node you send a GET to a practice CRM, filter it with a query parameter and POST a note back. Before that, Items, JSON and Expressions shows what the answer looks like inside n8n.
If you only need your own app data in a sheet, you may not need to write a request at all. A connector calls the API for you: in Pull Live Pipeline Data into Google Sheets with Coefficient, a sales pipeline lands in Sheets and refreshes daily with no code.
What are bearer tokens, pagination and rate limits?
These three cause most API workflows that "just stop working", and the course gives them their own module, Bearer Tokens, Pagination and Rate Limits, in Plain Words.
- Bearer token. An API key sent in the Authorization header as Bearer, a space and the key. It works for whoever holds it, so keep it out of URLs (they end up in logs and screenshots) and store it once as a credential.
- Pagination. API pagination means a big list comes back one page at a time with a cursor for the next one. If your sheet always has exactly 100 rows, you are only reading page one. n8n's HTTP Request node has a Pagination option that loops for you.
- Rate limits. A rate limit caps requests per minute. Go over and you get 429 Too Many Requests: wait and retry. A 401 means the key is wrong and a person has to fix it; a 500 means a problem on the API's side.
Pull or push: when do you need a webhook?
Everything above is pull: your workflow asks for data on a schedule. A webhook is push: the app sends the data to your workflow the moment something happens, such as a form being sent. Use it when speed matters, like a new lead. What a Webhook Is, in Plain Words shows one arriving.
How much do pay-per-call APIs cost?
Usually less than people expect. Work it out as calls per run, times runs per month, times the price per call. ProLevelMax listed its Subreddit feed at $0.003 per call in October 2026, so one call a day is about nine cents a month. A workflow that loops over thousands of items a day is a different bill, so multiply before you build, and check each provider's current pricing page. See API usage billing.
Two modules use paid data APIs in n8n. In Litescrape in n8n, a weekly competitor-price workflow moves from SerpApi to a cheaper search API by changing one HTTP Request node and moving the key into a Bearer Auth credential. In ProLevelMax in n8n, a Reddit summary workflow survives Reddit switching off its RSS feeds on November 13, 2026, by swapping the RSS Read node for one HTTP Request node.
What does the course cover, in order?
- How Stepthrough Works (two minutes)
- APIs vs Scraping vs RSS: the four ways in and the five parts of a request
- Coefficient: CRM data in Google Sheets, no code
- n8n items, JSON and expressions
- n8n HTTP Request node: your first GET and POST
- Bearer tokens, pagination and rate limits
- Webhooks: the API that calls you
- Litescrape: a pay-per-call search API
- ProLevelMax: replace Reddit RSS
- Error handling in n8n: get alerted when a source fails, and retry by itself
Key takeaways
- Official API first, RSS when a site offers it, a pay-per-call data API for search results and similar, scraping last.
- Every API call is a method, an endpoint, query parameters, headers with the key, and a response with a status code and JSON.
- Keep keys in saved credentials, follow pagination to the last page, and handle 401, 429 and 500 differently.
- Do the cost math before you build: calls per run times runs per month times the price.
- Build in steps, so when a source changes you swap one node, not the whole workflow.
Frequently asked questions
- Is web scraping legal?
It depends on the site, the data and where you are. Many sites forbid scraping in their terms of service, and personal data has extra rules. Prefer an official API or a data provider, and check the terms before you scrape anything. This is not legal advice.
- Do I need to code to use an API?
No. Connectors such as Coefficient make the API call for you, and n8n's HTTP Request node is a form: method, URL, parameters and a saved credential.
- What is the difference between an API and a webhook?
With an API, your workflow asks for data when it wants it. With a webhook, the app sends the data to your workflow the moment something happens.
- Is this course free?
Yes. All 10 modules are free and the practice needs no account. To build the workflows for real you need accounts in the tools each module names, and each module says what you need.
Sources
- n8n docs: HTTP Request node
- ProLevelMax pricing (read October 19, 2026, for the ProLevelMax module)
- Litescrape pricing (for the Litescrape module)