- What a feed address looks like
- Where sites put their feeds
- Common patterns by platform
- A page is not a feed
- When a site has no feed
What a feed address looks like
A feed address is an ordinary web address that answers with a feed document instead of a page: RSS or Atom, which are XML, or JSON Feed. Opened in a browser it shows raw markup or a plain list of posts, depending on the browser, and that is normal. A feed reader fetches that address on a schedule and shows the posts inside it.
Some examples of what feed addresses look like:
https://example.com/feed/
https://example.com/rss.xml
https://example.com/atom.xml
https://example.com/blog/feed.json
https://www.youtube.com/feeds/videos.xml?channel_id=UC...The address of a page, such as https://example.com/blog/, is not a feed address, even when the page lists the same posts. A reader given a page address looks for the feed the page declares, which is how the next section works.
Where sites put their feeds
Most sites name their feed in the page's head, in a link element that browsers and readers understand:
<link rel="alternate" type="application/rss+xml" title="Example" href="https://example.com/feed/">A feed reader given the page address reads this element and subscribes to the feed it names. That is what Inoreader does when you paste a site address into the subscribe box, and what the RSS feed finder does when you give it a page: it fetches the page, reads every declared feed and checks that each one is a feed that works.
Sites that do not declare their feed usually still have one at a conventional address, which is the second thing the finder tries.
Common patterns by platform
The platform a site runs on decides where its feed is. These are the usual addresses:
| Platform | Feed address |
|---|---|
| WordPress | /feed/, and /category/<name>/feed/ for one category |
| Blogger | /feeds/posts/default |
| Medium | https://medium.com/feed/@<user> or https://medium.com/feed/<publication> |
| Substack | https://<name>.substack.com/feed |
| Ghost | /rss/ |
| Tumblr | /rss |
| YouTube | https://www.youtube.com/feeds/videos.xml?channel_id=<id> |
https://www.reddit.com/r/<subreddit>/.rss |
|
| GitHub | https://github.com/<user>/<repo>/releases.atom, /commits/<branch>.atom |
| Mastodon | https://<instance>/@<user>.rss |
| Squarespace | /?format=rss |
| Shopify | /collections/all.atom |
A podcast's feed is the address the podcast directories list, usually on the show's page under "RSS" or in the directory's own page for the show.
A page is not a feed
The most common mistake is to subscribe to a page address and expect posts. A reader given a page that declares no feed has nothing to read; the validator reports it as "a web page, not a feed". Some sites answer their old feed address with a page too, after a redesign, and readers that were subscribed stop updating without any error the publisher sees. The RSS feed validator tells the two apart, and for a page it lists the feeds the page declares.
When a site has no feed
Many sites have a feed and do not show it; the finder finds those. A site with no feed at all can still be followed in Inoreader with a web feed, which watches the page for new entries and turns them into a feed. The web feeds feature page explains how. Newsletters have their own way in: an Inoreader email address subscribed to the newsletter turns each issue into an item.