Most SEO news does not matter. The trick is not reading more of it. It is knowing which slice changes what you do on Monday. Right now that slice all points one way: search is turning into an answer.
How to read an algorithm update without panicking
Google confirms a core update, the SEO feeds melt down, and half the advice that week is written by people watching a rank tracker refresh. Here is the calmer version.
Wait for the rollout to finish before you conclude anything. Core updates take two to three weeks, and judging your traffic on day three is like weighing yourself halfway through a meal. When it settles, segment the change before you react. A drop concentrated in one folder or one type of query is a specific, fixable problem. A smooth decline across the whole site is usually a bigger and slower one.
Whatever you do, change nothing while the update is still rolling. Ship five fixes mid-rollout and you will never know which one mattered, or whether the page would have recovered on its own.
Core updates rarely "penalise" a site. They re-score what Google already believed about quality across the entire index. Recovery means genuinely improving the pages that fell, then waiting for the next update to look again. There is no button.
Search is becoming an answer, and that changes the job
AI Overviews, ChatGPT search, Perplexity, Google's AI Mode: different products, same behaviour. They read several pages, write one answer, and the user often never clicks.
So being cited now matters as much as ranking. A mention inside the AI answer can send qualified traffic even when your blue link sits at position eight. At the same time, informational queries are bleeding clicks while commercial ones hold: "what is X" gets answered in place, but "best X for Y" and "X vs Y" still send people to pages because buyers want to verify before they spend.
The third shift is subtler. Brand recognition has become a ranking input. Models lean toward names they already associate with a topic, so being a known entity is not a vanity goal. It is how you get retrieved in the first place.
E-E-A-T, translated into things you can do
Experience, Expertise, Authoritativeness and Trust is not a dial in the algorithm. It is a set of signals human raters look for and Google's systems try to approximate. Your job is to make those signals obvious.
That means real, named authors with a bio and a track record instead of "Admin" or "Editorial Team". It means first-hand proof: your own screenshots, your own test results, your own numbers. It means linking primary sources, dating your statistics, and only making claims you can stand behind. And it means a findable business behind the site, with an about page, real contact details and a stated editorial process.
Keep a changelog. One shared doc listing what you shipped, on which pages, on which date. When traffic moves, you will have cause and effect instead of a group chat full of guesses.
What we're watching in 2026
Four questions are shaping the year: how far AI Overviews cut into informational traffic and which content formats survive it, whether llms.txt becomes a real standard or fades like humans.txt did, how much of both classic and AI results Reddit and YouTube keep taking, and how fast the SEO tool market consolidates around AI-visibility tracking.
Google algorithm updates: a short 2024 to 2026 timeline
You do not need to memorise update names. You need the pattern: Google now ships large core updates several times a year, each rolling out over two to three weeks, each re-scoring quality across the whole index.
| Update | When | What it targeted |
|---|---|---|
| March 2024 core and spam | Mar 2024 | Scaled content abuse, expired-domain abuse, parasite SEO; a large reduction in low-quality content |
| Aug, Nov, Dec 2024 core | 2024 | Rewarding smaller independent sites, adjusting earlier over-corrections |
| March 2025 core | Mar 2025 | Continued quality re-scoring; helpful-content signals folded into core |
| June to Nov 2025 core | 2025 | Ongoing, with visible volatility for review and affiliate content |
| 2026 core updates | ongoing | Same direction: originality, first-hand experience, brand trust |
The recovery playbook has not changed: wait for the rollout to finish, segment the drop by folder and query type, genuinely improve the pages that fell, then wait for the next core update to re-evaluate them. There is no button.
Zero-click search: how much traffic is really disappearing?
A zero-click search ends without the user visiting any website, because the answer is on the results page: an AI Overview, a featured snippet, a knowledge panel or the map pack. The share has climbed, but it is uneven, and commercial intent is largely intact.
Figures vary by study and industry. The pattern is stable: the closer to a purchase, the more the click survives.
The response is not to panic, it is to rebalance. Shift more of your content mix toward comparison, "best", "vs" and pricing pages, which still earn the visit. Treat AI Overview and snippet citations as a branding channel, and start measuring the referral traffic assistants send you. Build direct channels like email and community so the click is not your only lifeline. And report impressions and citations right next to clicks, so a page that gets summarised does not read as a failure on the dashboard.
Content decay: why traffic falls, and the refresh cadence that fixes it
Content decay is the slow decline in traffic to a page that once ranked, as competitors update, intent drifts and the information ages. It is normal. On most established sites, a planned refresh programme recovers more traffic than new publishing does.
A simple quarterly routine keeps it in check:
- In Search Console, compare the last three months to the previous three, and list every page down more than 20%.
- For each one, check what now outranks you and what changed. Update the facts, add what is missing, sharpen the intro, refresh the examples and the date.
- Resubmit the URL and add one or two fresh internal links from strong pages.
- Retire or merge pages that have lost their query entirely, with a 301 to the closest match.
Illustrative example
Refreshing beats publishing on an ageing blog
Imagine a five-year-old blog with around 300 articles. New posts land with a thud because the site is spread thin, and total organic traffic has been drifting down for a year.
Instead of publishing anything new for a quarter, the team refreshes the 40 pages that lost the most traffic: updated data, tighter intros, new sections answering questions that now appear in the results, fresh internal links.
A focused refresh cycle like this usually recovers more traffic than three months of new publishing would have, at a fraction of the effort.
E-E-A-T in 2026: which trust signals correlate with rankings?
E-E-A-T, for Experience, Expertise, Authoritativeness and Trust, is not a score in the algorithm. It is a framework Google's raters use and its systems approximate. After the 2024 and 2025 updates, the signals that consistently line up with recovery and resilience are concrete.
| Signal | Weak version | Strong version |
|---|---|---|
| Authorship | "Admin" or "Editorial Team" | Named author, real bio, credentials, track record |
| Experience | Rewritten from other articles | First-hand tests, own screenshots, own data |
| Sourcing | Vague claims, no links | Primary sources linked, statistics dated |
| Business identity | No about or contact page | Full about page, real contact details, editorial policy |
| Reputation | No third-party presence | Reviews, mentions and citations on trusted sites |
- Every article has a real, named author with a linked bio
- Money and YMYL pages show a reviewer or editor and a last-updated date
- Claims link to primary sources; statistics carry a date
- The site has an about page, contact details and a stated editorial process
- Your brand is mentioned and reviewed on sites your audience already trusts
How to stay current in 20 minutes a week
Pick three or four practitioners who publish tests, not takes. Check Google's Search Status Dashboard only when you see movement. Skim one weekly roundup. Ignore the rest. The people who obsess over daily volatility are rarely the ones with the best rankings.
