How to use this site

Everything here in one place, from the five-minute version to the parts most people never find. If you only read one section, make it the workflows at the bottom — they are what the rest of the site exists to serve.

1. Start here

This site reads 46 feeds — Google's own channels, the search press, specialist practitioners, and the adjacent disciplines most SEO sites ignore — and sorts them so you can see what matters without opening a dozen tabs.

Three things are worth knowing before anything else:

The lead story is the most consequential recent item, not the newest. Whatever published last does not automatically win the top slot.
Everything links out. We summarise and point at the publisher who wrote it. Nothing is republished here.
The front page is search-focused. The UX, conversion and marketing feeds are deliberately kept off it so the daily search news stays readable — browse those sections directly, or search across everything at once.

2. Reading the front page

High impact: Likely to affect a single website — a confirmed ranking update, a policy change, something demanding action. Roughly 4% of stories. Medium is worth reading; unbadged stories are industry background. How the score works.
Update: Google confirmed a ranking change. These also feed the Update Radar strip.
4 outlets: Several publications covered the same story. We show it once and name the rest underneath — a story four outlets ran is usually a story that matters.
Crawling & Indexing: The technical area a story touches. There are 15 of them; click one to see everything tagged with it.
Update Radar: Confirmed algorithm updates pulled to the top, whatever page you are on.
Model Watch: New AI model releases from the labs behind AI search. First-party announcements only — which model sits behind an assistant changes what it retrieves and who it cites.

3. Narrowing it down

The navigation has two rows, and the split is deliberate:

Search — Google's own announcements, other engines, algorithm and industry reporting, AI search, specialist practitioners, research and tools.
Wider craft — UX, conversion and analytics, and the broader marketing mix. Rankings get someone to the page; these decide what happens next.

Below the toolbar there are two more filters that stack with each other:

Impact on your site — High, Medium or Low, with live counts. Clicking an active chip again clears it.
Technical area — Ranking, Crawling & Indexing, Site Architecture, Structured Data, Core Web Vitals, Links, Local, E-commerce, Policy, AI Search and more.

Try this: High impact + Policy & Penalties is the closest thing here to "things that could get my site in trouble".

4. Searching properly

Search covers every source, including the sections kept off the front page. Word order never matters.

Type this	To get
`core update`	Both words, in any order, anywhere in the story
`"core update"`	That exact phrase only
`core -spam`	Has `core`, does not mention `spam`
`"ai overviews" -chatgpt`	Operators combine freely

Results rank by relevance — a headline match counts for far more than one buried in a summary — and you can switch to newest-first. Six filters narrow further: section, source, topic, impact, date range and sort.

Try this: crawl budget · "ai overviews" · schema -video

5. Making it yours

The toolbar above the stream holds five things. All of them remember your choices in your own browser — nothing is sent anywhere, and nothing is tied to an account, because there are no accounts.

New since your last visit: Stories published since you last looked are flagged NEW with a count at the top. Refreshing does not clear them — the marker only moves when a genuinely new visit starts. Mark all read resets it deliberately.
Compact: One line per story instead of cards. Roughly three times as many headlines per screen; the fastest way to scan a busy day.
Watchlist: Add terms — a client brand, local, schema, a competitor — and matching stories are highlighted and counted. Tick show only watched to hide everything else.
Sources: Hide publishers you never want to see, permanently.
Keys: Keyboard shortcuts, listed below.

Key	Does
`j` / `k`	Next / previous story
`o` or `Enter`	Open the selected story
`/`	Jump to the search box
`c`	Toggle compact view
`m`	Mark all read
`?`	Show the shortcut list
`Esc`	Close panels, clear selection

These five need JavaScript. Everything else on the site — reading, filtering, searching, every export — works without it.

6. Taking it with you

You are not required to read any of this on our site. Four ways to take it elsewhere:

OPML — all 46 feeds

One file, importable into Feedly, Inoreader, NetNewsWire or any reader. Feeds arrive nested under their sections, so the structure survives the import instead of landing as one flat pile.

Every feed here was checked by hand for validity and for whether it is still publishing. That checking is most of the value.

Combined RSS

Everything the front page shows, in one feed — search-focused, with the lab and adjacent-discipline noise already filtered out.

Updates-only RSS

Confirmed Google ranking updates and nothing else. Point a Slack incoming webhook at it and you get told when an update starts and again when it finishes, without thirty headlines a day.

The alert channel most people actually want.

Update archive — CSV / JSON

The whole Google update history as a file, with computed rollout durations. Drops straight into a client deck or a Looker Studio annotation layer.

Carries Data source, Date precision and Name origin columns, so you can tell which rows came from Google's API and which were compiled.

7. The reference pages

Google update history: Every core update, spam update and ranking incident, with rollout durations Google never publishes — it gives start and end times but never the elapsed figure. Includes a date lookup that answers "was Google updating on the 14th?". There is a plain-English explainer on that page covering what core and spam updates actually are and what to do when one lands.
Recaps: Weekly, monthly and yearly summaries built from our own archive: the highest-impact stories, which updates landed, and which technical areas were busiest.
Every source we follow: All 46 feeds grouped by discipline, with what each is good for and when it last published. If you are new to the field, reading down that page is a faster orientation than any "best SEO blogs" list.

8. Four workflows

The parts worth stealing.

The two-minute morning check

Open the front page. Read the new since your last visit count.
Press c for compact — three times the headlines per screen.
Scan the Update Radar. If it is empty, Google confirmed nothing.
Read only High impact and anything your watchlist highlighted.
Press m to mark everything read, so tomorrow starts clean.

Diagnosing a traffic drop

Get the date the drop starts from your analytics.
Put it into the date lookup. It tells you which updates were live that day and how far into the rollout you were — day 3 of 12 means very different things from day 12 of 12.
If something was rolling out: wait for it to finish before changing anything. Mid-rollout data is noise.
If nothing was: the cause is more likely yours — a deploy, a robots.txt change, a migration, a server problem.
Either way, search that window for coverage: core update, past month.

Tracking a client or a vertical

Open Watchlist and add the client's brand, their category, and any platform they depend on.
Add the technical areas that matter for them — a large e-commerce site cares about Crawling & Indexing; a multi-location business about Local SEO.
Mute publishers you never act on, so the highlights stand out.
Check the watchlist count each morning; open the panel only when it is non-zero.

Building a monthly client report

Open the monthly recap for the period.
Take the confirmed updates and the highest-impact stories — that is your "what changed in search" section written for you.
Download the update CSV and use the start and end dates as annotations on your traffic charts. Rollout windows explain far more chart movement than most reports admit.
Use the busiest-technical-areas chart to justify what you prioritised.

Something missing?

If a feature here is unclear, or you want a source added, the address is on the contact section. Suggestions for feeds we should be reading are genuinely welcome — every current source was added the same way, by someone deciding it was worth checking.