Blog RSS Feeds Explained: How They Work for Newsletters
Newsletter Creation

Blog RSS Feeds Explained: How They Work for Newsletters

You're three stories short of a full issue and the send is tomorrow morning. Twelve tabs are open: two competitor blogs, a subreddit, a YouTube channel, and a handful of industry sites you check by hand every week. Somewhere in that pile are the three items your readers will care about.

The file that can do most of that checking has been sitting on nearly every blog since the late 1990s. It's small, unglamorous, and most creators never open it. That file is the blog RSS feed, and once you know what's inside it, sourcing stops being a tab problem and becomes a pipeline.

Here's the path from that file to a delivered issue: what a feed contains, how to find one, why feeds misbehave, and how a stack of them becomes a shortlist you can write from.

What a blog RSS feed is and why creators still use one

A blog RSS feed is a standard XML file that lists a blog's newest posts, so a reader app or newsletter tool can check for updates without loading the site. Everything else about feeds follows from that one sentence.

The easiest comparison is a table of contents that rewrites itself every time the blog publishes. Instead of you opening twelve sites to see what changed, one piece of software reads twelve small files and reports back. The format was built for machines from the start: Netscape released RSS 0.90 in March 1999 to expose headlines for its My.Netscape.com portal, as the history of RSS on Wikipedia records, and the later 0.91 and 1.0 branches kept that machine-to-machine purpose intact.

Why this matters for a newsletter: a feed is the cleanest source list an issue can read, because every new post arrives with a title, a link and a date already attached. No scraping a homepage, and no guessing when something went live.

Plenty of people and tools still poll those files. One personal blog's feed took 537,541 requests between January 1 and September 1, 2025, according to the blog's publisher iDiallo. Some of that traffic comes from readers and some from software. Both point the same way: the file is being pulled constantly, so it belongs in your process rather than gathering dust in your bookmarks.

What sits inside that file is where the practical decisions start.

The parts of a feed creators should recognize

You can judge a source in about 30 seconds once you know which fields to look at. Every feed has two levels: a channel that describes the whole blog, and item entries that each describe one post, as the RSS 2.0 specification published by Harvard's Berkman Klein Center sets out.

The structure is simple enough to hold in your head, and it explains every symptom you'll meet later.

A diagram explaining the structure of a blog RSS feed, showing the Feed, Channel, and Items hierarchy.

The channel tells you what you're subscribing to

Channel metadata covers the feed title, description, publication date, language and update information. In practice the title should match the blog you meant to follow and the description should tell you the beat. Language matters more than people expect, because a feed in a language your readers don't speak is a source you'll never use no matter how good the posts are.

A generic channel title is a useful warning sign. When a blog's feed names itself after its content system rather than its publication, the address may be a default output rather than the feed the publisher intended.

The item level is where issues get built

Each item can carry a title, description, link, author, category, publication date, enclosure and a globally unique identifier, usually written as guid. Those elements are individually optional, but the specification holds one hard rule: an item must contain at least a title or a description. A title plus a link plus a description is what gives you enough to show a story and credit the source.

Element Required Newsletter use
title At least one of title or description Headline for the story card and subject line ideas
description At least one of title or description Summary or full HTML body, sometimes only an excerpt
link Optional Canonical link back to the original post for credit
guid Optional Stable ID that lets tools spot duplicates
pubDate Optional Publication date for recency filtering and sorting
author Optional Byline for credit lines in roundups
category Optional Topic tags for grouping stories
enclosure Optional Attached media such as an audio file

One detail trips up most pipelines: RSS lets a description hold either a short synopsis or entity-encoded full HTML. A tool reading that field has to sanitize the markup, pull out readable text, and avoid assuming the full article is present, as the RSS 2.0.1 specification notes collected by RSSboard make clear.

A channel can also hold zero items and remain perfectly valid, which is worth remembering the next time a feed looks empty during a quiet publishing week. The guid is the field that makes duplicate detection possible, and it causes most of the strange behaviour discussed further down.

What a title and summary owe your reader

Those two fields carry more weight than they look like they do. Mozilla's accessibility guidance on the ARIA feed role describes article-by-article browsing for screen reader and keyboard users, which only holds up when each entry has a clear title and a summary that says something. A bare teaser gives assistive technology nothing to decide on.

On the reuse side, keep the habit simple. Quote a headline, link back to the original, and summarize rather than republish full text. If you plan to run long passages from someone's posts in your issue, read their stated reuse terms first and ask when the terms aren't clear.

How to find and add a blog's feed

The address you need is the feed URL itself. Paste a blog's homepage into a reader and you'll get either an error or a formatted page the tool can't parse.

An infographic showing a four-step process on how to find and add an RSS feed to a reader.

  1. Open the blog and look for a feed icon. An RSS symbol or a link labelled feed in the header, sidebar or footer is the fastest route. When it's visible, the publisher has already done the work for you.

  2. Try the two common suffixes. If there's no visible link, add /feed or /rss to the domain and load it. You're looking for a page of raw XML text rather than a designed page with images, which confirms you've found the machine version and not a styled archive.

  3. Copy the full feed address from the address bar. Copying a shortened link from an email or a chat thread is how feeds end up loading once and failing later. The full URL is the one that keeps working.

  4. Check the three newest items. Confirm a title, a working link and a publication date on each one. Those are the three fields every later step depends on, and a source that fails this check will fail again when you're assembling an issue on deadline.

  5. Group the feed by purpose. Keep competitor blogs in one group and industry research in another, so an issue draws from a defined pool instead of whatever you happen to remember at 9pm.

If /feed and /rss both come back empty, the blog may publish under a tag, a category or a subdomain. In that case, open an individual post and look for a feed link there. Some sites also block automated checks of their pages, which is why a direct feed URL is often the only reliable route in, and why we let writers paste a feed link directly, including feeds on sites that block automated checks. Subscribing to any public feed in HeyNews brings its new articles in automatically, so the paste is the whole setup step.

Which sources deserve a slot is its own decision. This guide to news monitoring tools compares the options when you're weighing a reader against a full monitoring setup.

Why feeds duplicate, vanish, or go quiet

A feed can hand your draft the same story twice, skip a post entirely, or stop delivering without ever throwing an error. Almost all of that trouble traces back to two fields: the identifier and the date. Feed readers generally treat items that share the same guid as the same story, so the moment a publisher regenerates that identifier, a repeat starts to look like a brand new item in your pipeline.

The duplicate problem starts at the publisher

When a writer edits a post and the platform regenerates the identifier, your tool sees a new story. Same headline, same link, two slots in your draft. The same thing happens when a post's URL changes: a permanent link can stand in as the guid, and an unstable or time-based one can't. Nothing in the item itself marks the second appearance as an edit, which is why deduplication has to happen on your side before anything reaches a pick list.

Practical rule: dedupe on identity first. Comparing headlines and timestamps to catch the same post is guesswork, and it breaks the moment an editor rewrites a title.

Symptom Likely cause Fix
Same post appears twice GUID or URL changed on an edit Dedupe on GUID and canonical link, and ask the publisher to keep IDs permanent
Older post is missing Feed lists only the latest items Check for paging or archive feeds before assuming the post is gone
Item has no date Publisher omitted pubDate Use fetch time and flag the story as date uncertain
Feed will not load or shows errors Malformed XML or a homepage URL Validate the feed and paste the feed address itself
Summaries too thin to rank Description holds a teaser only Open the link and score the full article text

Every one of these failures lands in the same place: the shortlist your next issue is built from.

A missing post has a similar explanation. Many feeds publish only the latest batch of items, while older entries sit in paging or archive documents defined by RFC 5005 on feed paging and archiving. A post that vanished from page one may still be sitting one document over.

Two more habits keep a stack honest. Treat a publication date as a fact about when something was posted, not proof that its claims are current. With RSS seeing renewed interest amid concerns over social media and privacy, more feeds flow into the average stack, so flag evergreen-looking items for a check before they run. Watch payload size too: Google News publisher guidance caps an article at 1 MB and a feed at 2 MB, per the Google News publisher help page on feed size, a useful ceiling even when your own tool accepts more.

Once a week, open each feed and count duplicates, missing dates and failed loads. Drop any source that fails the same check twice in a row. A stack of twenty clean sources beats sixty unreliable ones every time.

How creators turn feeds into a newsletter issue

Twelve sources can produce a few dozen items in a week, and an issue needs five. Everything after collection is a decision about which five, and why.

Say you follow twelve sources: two industry blogs, a subreddit, a YouTube channel, several newsletters and a couple of competitor sites. In a single week those feeds return a few dozen new items. That number illustrates the shape of the workload rather than any measured average. Feed reading is still how a lot of publishers work, and in 2025 TechRBun reported that more than 60% of independent bloggers still actively used RSS readers to track industry news and competitor updates.

Five stages from raw items to a pick list

  1. Collect. Pull new items from every feed since your last send. Keep the raw title, link, date and identifier for each one.

  2. Dedupe. Remove repeats by guid or canonical link before anything else touches the list, so you're scoring real stories rather than the same post twice.

  3. Filter by window. Cut anything older than the issue window. A source that updates slowly shouldn't push last month's news into this week's slot.

  4. Score for fit. Rank the survivors against the topics your readers signed up for. A story that's huge in another industry is noise in yours, and fit beats buzz every week.

  5. Choose and angle. Pick the strongest five and write one line on why each earns its slot. That line becomes the paragraph that carries the item.

A person organizing content from twelve various online sources into a curated weekly newsletter report.

Stages four and five are where tooling earns its keep, because they're the ones that need judgement applied at speed. HeyNews checks the feeds a writer follows on a schedule, scores new stories for that audience and drafts the issue in the publication's voice, from a source pool the writer controls. Every draft waits for review, and edits happen before anything leaves the app.

The creator still decides what runs. A ranked list is a suggestion. The editor's judgement about what belongs in front of readers is the part that can't be handed over.

The handoff between sourcing and drafting is where most of the week disappears, which is why it's worth mapping the newsletter content curation workflow before you add another source to the pile.

Set up your first feed stack this week

Pick five feeds you already read and run one real issue through them before you add a sixth. That test costs an afternoon and shows exactly where the stack leaks before the leak reaches readers.

  1. Confirm each feed loads. Titles, links and dates on the three newest items, every time. Drop any source that fails.

  2. Collect a week of items. Log every duplicate, missing date and thin summary. Two feeds republishing the same story is the duplication problem in miniature, and a simple log makes the pattern visible fast.

  3. Draft one issue by hand from that material. Count the minutes spent stripping repeated stories, chasing dates and expanding thin summaries. That figure is your real per issue cleanup cost, and it decides which sources earn a permanent slot.

  4. Fix the worst source. Replace it, or ask its publisher to keep identifiers permanent so the same post stops resurfacing as new. When you're ready to automate the scan, any feed, site, subreddit, YouTube channel or social profile can be added as a source in one paste, and each new story arrives on a schedule with a relevance score.

This test works with or without a tool. For the arithmetic behind doing it by hand, the breakdown of source monitoring hours per year is worth ten minutes.

A feed is only as useful as its identifiers and dates. Start with the five sources you'd have opened tomorrow morning anyway.


HeyNews follows the RSS feeds you read, scores new stories for your audience, and drafts the next issue in your publication's voice. You paste a feed link, it checks on a schedule, and you get a draft to edit rather than a blank page on deadline. Every issue still waits for your review, and you send from the platform you already use. Open app.heynews.co, paste your first feed link, and see what your own source list turns into.

Cagri Sarigoz, Co-founder of HeyNews

Cagri Sarigoz

Co-founder & CEO of HeyNews. Cagri has spent 15+ years in growth and technical marketing, mostly figuring out how to make AI do the tedious parts of content creation so humans can focus on the interesting parts. At HeyNews, he builds the systems that turn RSS feeds, Reddit threads, and blog posts into a newsletter that sounds like you wrote it.

Try the AI newsletter generator that sounds like you

Start Free Trial

See how it works