Collect Medium article data through one API request
Send a Medium URL and an API key, and receive parsed article data without running your own browser fleet, proxy rotation or anti-bot handling.
Each call can return fresh page data ready for monitoring, trend analysis or research workflows.
How the request flow works
Use this workflow when you need repeatable Medium data rather than one-off manual copying. The steps below describe the intended request flow; confirm current account terms and SDK availability before relying on them in production. First, obtain an API key through the service's signup process. Second, install the SDK for your stack or use plain HTTP/cURL; Python, Node, Go, PHP, Ruby and Java are listed as supported options. Third, send one request containing your API key and a Medium URL, with proxy rotation, fingerprint handling and JavaScript rendering managed behind the scenes. Fourth, add advanced options only when needed: JavaScript rendering for dynamic content, stealth options for CAPTCHA-heavy pages, country targeting for geography, and AI extraction rules for natural-language field selection. As a practical check before scaling, run one article URL and confirm that title, author, claps, response count, tags, publication, reading time and body text are present, then test a second publication page. If fields are missing, adjust extraction rules rather than adding more requests.
What to collect, by use case
Developers typically use Medium data for three distinct jobs, and the right fields depend on which job you are doing.
Track publication output: follow specific writers, publications and tags, and pull headlines, authors, claps and response counts to see what is being published and how it is resonating.
Monitor content trends: watch emerging topics across publications to inform strategic and competitive decisions. Engagement fields such as claps and responses help separate widely read pieces from low-engagement posts.
Extract author names: identify who wrote technical or specialist pieces, which can support collaborator discovery, research partnerships and idea sharing.
For brand-mention monitoring, combine the author and publication fields with body text so each record stays attributable to a writer and outlet.
Fields and output formats
Every Medium page can return a consistent field set: article title, author, claps, response count, tags, publication, reading time and body text. That consistency is what makes the output queryable downstream instead of a pile of HTML.
Two extraction paths lead to the same structured JSON result. With CSS selectors, you map each selector to a named field, which gives precise control when page structure is stable. With AI extraction rules, you describe the fields you want in natural language, which is faster to set up when selectors would be brittle. Screenshots and full-page capture are also available when you need visual verification alongside the parsed fields.
Additional outputs can be useful in specific pipelines: clean Markdown for feeding articles into summarisation or trend-analysis workflows, JavaScript scenarios that click, scroll or type before extraction, and no-code automation through n8n for content-monitoring workflows without writing scraper code.
Why use an API instead of manual collection
Manual collection is slow, hard to scale and limited by anti-bot defences. The API approach shifts browser management, proxy rotation and rendering to the service, so your team focuses on the parsed data.
Key capabilities described for this workflow include JavaScript execution, proxy rotation without straining your own servers, and headless-browser rendering that mimics real browsing behaviour to reduce blocking risk. The platform is described as scaling across thousands of browsers and rotating proxies, with typical results in 1-5 seconds even on JavaScript-heavy pages.
Compliance note: the service states it does not collect or store personal data from scraped sites unless a user explicitly requests it, and describes itself as GDPR and CCPA compliant. Regardless of tooling, review Medium's terms of service before deploying and avoid post-login content.
Pricing and credit planning
Before choosing a plan, confirm the current pricing, included usage, trial conditions, billing schedule and cancellation terms.
Before choosing a tier, estimate credit use per request rather than per page alone. Requests start at 1 credit each, JavaScript rendering adds cost, premium proxies cost more depending on whether JavaScript is enabled, and stealth proxy options are the most expensive per request. Credits are described as valid for one month without rollover, and concurrency limits determine how many requests can run at the same time.
A simple sizing check: multiply your expected URLs by the credit cost of the options you actually need, then compare against the monthly credit allowance and concurrency limit of each tier. Start with the free allowance to validate field quality before committing to a paid plan.
| Field | Use it for |
|---|---|
| Title | Identifying and deduplicating articles |
| Author | Tracking writers and finding collaborators |
| Claps | Measuring reader engagement |
| Response count | Comparing discussion depth across pieces |
| Tags | Grouping content by topic |
| Publication | Attributing output to an outlet |
| Reading time | Estimating content length at scale |
| Body text | Full-text analysis and LLM workflows |
Fields described as returned for every Medium page.
Frequently asked questions
Do I need an account to scrape Medium?
No. You only need an API key from this service. The free-credit allowance and any card requirement are set by the current account terms, so check the signup page for the applicable figures before you start.
Is scraping Medium legal?
Public-facing pages can usually be collected, but you should review the target site's terms before deploying and avoid post-login content.
What data can I extract from Medium?
Anything rendered on the public page can be targeted. Use CSS-based extraction rules or AI-driven rules to return structured JSON.
How do I get structured JSON from Medium?
Map CSS selectors to field names with extraction rules, or describe the fields in natural language with AI extraction rules. The parsed response is returned as JSON.
Which programming languages can I use to scrape Medium?
Any common stack works. SDKs are described for Python, Node.js, Go, PHP, Ruby and Java, and plain cURL also works.
How much does it cost to scrape Medium?
Credit use varies by option: JavaScript rendering, premium proxy and stealth proxy options carry different per-request costs. Check the current credit table before estimating volume, and confirm plan prices, included usage and billing terms with the provider.