Few things create panic in SEO quite like opening Search Console and seeing traffic falling at the same time Google announces an algorithm update.
Rankings are moving.
Clicks are down.
Slack messages start arriving.
Someone asks, “Have we been hit by the update?“
And suddenly everyone wants to change something.
I’ve learned that this is usually the moment when doing less can be more useful than doing more.
Google updates are important. They can significantly change organic performance. But reacting too quickly can make it harder to understand what actually happened.
Before changing titles, rewriting pages or deleting half the website, I want to know three things:
What type of update was it? What actually changed? And do we have enough evidence to say the update caused it?
Those questions matter because not every Google update is the same.
First: Not Every Google Update Means the Same Thing
Google has already given us several useful examples in 2026.
As of September 2026, Google’s Search Status Dashboard shows a February Discover update, spam updates in March, June and August, and core updates in March and May. (Google Search Status Dashboard)
That’s important because I wouldn’t investigate all of those updates in exactly the same way.
February 2026 Discover Core Update
Google started rolling out a Discover core update on February 5 for English-language users in the US. The rollout was completed on February 27. Google described it as an update designed to improve the overall quality of Discover.
That’s already different from a normal Search ranking update.
If my traditional web search traffic remained stable but Discover traffic changed dramatically during that period, I’d start my investigation in Discover, not immediately conclude that the entire site’s SEO had been hit.
March 2026 Spam Update
On March 24, Google launched a spam update globally and across all languages. It completed on March 25. (March 2026 spam update)
A spam update deserves a different line of questioning.
Google says spam updates are improvements to its automated systems for detecting search spam. If a site loses visibility following one, Google’s own advice is to review its spam policies.
Now I’m thinking about things like scaled content abuse, scraped content, manipulative links, expired-domain abuse, site reputation abuse and other practices covered by Google’s spam policies.
That’s very different from simply asking whether an article needs another 500 words.
March 2026 Core Update
Three days after the spam update began, Google launched the March core update on March 27. It finished rolling out on April 8. (March 2026 core update)
That overlap matters.
Imagine traffic dropping around March 28.
Was it the spam update?
The core update?
Something you changed on the website?
Seasonality?
A competitor improving?
A tracking problem?
That’s exactly why I’m careful when someone immediately says:
“The Google update caused this.”
Correlation gives us somewhere to investigate.
It isn’t automatically the diagnosis.
May 2026 Core Update
Google launched another core update on May 21, completing the rollout on June 2. (May 2026 core update)
Core updates are broad changes to Google’s ranking systems. Google specifically says they aren’t designed to target individual pages or websites.
That’s an important distinction.
If a site drops after a core update, I wouldn’t approach the investigation as:
“What did we do wrong?”
I’d start with:
“What does Google appear to be rewarding differently?”
Those aren’t the same question.
June and August 2026 Spam Updates
Google then launched another global spam update on June 24, completing it on June 26. (June 2026 spam update)
Then came another one on August 18.
The August 2026 spam update completed on August 21, making it the third confirmed spam update of the year. (August 2026 spam update)
So already, 2026 gives us a useful lesson:
“Google update” isn’t a diagnosis. It’s a category.
You need to understand which update happened before deciding how to respond.
My First Rule During an Update: Don’t Panic
This sounds obvious until rankings start disappearing.
Imagine one of your most important keywords moves from position three to position nine.
Then position eleven.
Then five.
Then fourteen.
All within a few days.
The instinct is to fix it.
Change the title.
Rewrite the introduction.
Add FAQs.
Change the H1.
Build links.
Add another 1,000 words.
But if Google is still recalculating rankings, what exactly are you fixing?
You don’t know yet.
Google’s own core-update guidance recommends waiting until a core update has finished rolling out before analysing the impact. It suggests waiting at least a full week after completion before comparing performance.
That makes sense to me.
During the rollout, monitor.
After the rollout, analyse.
Then decide whether you need to act.
Before Blaming Google, Check Whether You Broke Something
This is one of the least exciting parts of SEO.
It’s also one of the most important.
A Google update starts on Monday.
Traffic drops on Tuesday.
Everyone blames Google.
Then on Thursday, someone discovers a developer accidentally added noindex to an important template.
Google didn’t do that.
Before diagnosing an algorithm hit, I want to rule out the boring explanations.
Did tracking break?
Did the website go down?
Did robots.txt change?
Did canonicals change?
Did important pages become noindexed?
Were redirects implemented?
Was there a migration?
Did navigation change?
Were internal links removed?
Did the CMS update?
Did URLs change?
Did Google stop crawling important sections?
Was there a Search Console reporting issue?
Did seasonality change demand?
Did paid campaigns stop and alter other behaviour?
Did something happen to the brand?
Sometimes the best algorithm analysis begins with proving that the algorithm isn’t the problem.
Establish the Timeline
This is where I start treating SEO more like an investigation.
Let’s say Google launches an update on August 18.
Your traffic decline starts on August 3.
That’s useful information.
The August update probably isn’t a very convincing explanation for something that started two weeks earlier.
But suppose rankings are relatively stable until August 18, and then visibility starts falling across several sections of the website.
Now we have something worth investigating.
I want to map:
Update begins → volatility begins → affected pages/queries → update ends → performance stabilises.
The closer those events align, the stronger the case becomes.
Still not proof.
But better evidence.
Don’t Look at Sitewide Traffic First
This is probably one of the biggest mistakes I see in SEO analysis.
“Organic traffic dropped 18%.”
Okay.
Where?
The homepage?
Blog?
Product pages?
Service pages?
Locations?
One country?
One device?
Branded searches?
Non-branded searches?
Ten pages responsible for most of the decline?
You need to keep slicing the data until the problem becomes specific enough to investigate.
A 20% website decline could turn out to be:
one article that previously generated enormous traffic losing rankings.
That’s a very different problem from:
80% of commercial pages dropping across hundreds of keywords.
Sitewide averages hide stories.
Go to Search Console and Find the Actual Loss
This is where Search Console becomes much more useful than staring at an overall traffic graph.
I want to compare performance before and after the update.
Not just clicks.
Look at:
- clicks;
- impressions;
- CTR;
- average position;
- queries;
- pages;
- countries;
- devices;
- search appearance.
Google itself recommends reviewing top pages and queries before and after a core update and distinguishing between small and large ranking changes.
If a page moved from position two to four, I’m probably not rebuilding it.
If it moved from four to 29 and stayed there?
Now I want to understand why.
Find the Pattern Before You Find the Solution
This is the part I enjoy most.
Suppose 100 pages lost traffic.
Don’t immediately audit all 100 independently.
Ask what they have in common.
Maybe they’re all informational articles.
Maybe they’re all location pages.
Maybe they’re all thin category pages.
Maybe they’re all AI-assisted content published during the same period.
Maybe they’re old content.
Maybe they’re pages with weak internal linking.
Maybe they’re all targeting topics where Google increasingly prefers forums.
Maybe competitors with stronger first-hand experience replaced them.
Maybe they’re programmatically generated pages.
Maybe they’re all pages sitting three clicks deeper than the pages that remained stable.
That’s where an algorithm update becomes useful.
It can expose weaknesses that were already there.
Don’t Just Study What Lost. Study What Won.
This is probably the most important part of update analysis.
If your page dropped from position three to twelve, somebody replaced you.
Who?
Open the SERP.
Look at the pages that moved above you.
What changed?
Did Reddit appear?
YouTube?
Forums?
More authoritative brands?
More specialised websites?
Pages with original research?
Product pages instead of informational articles?
Local results?
AI features?
Different search intent altogether?
Maybe your page didn’t suddenly become terrible.
Maybe Google changed what it believes satisfies that query.
That’s a completely different problem.
I think SEOs sometimes spend too much time staring at the losing website.
The SERP is telling you what won.
Study that too.
Core Update? Think Quality and Relevance Before “Fixes”
For a core update, I wouldn’t start with tricks.
Google’s guidance is centred around creating helpful, reliable, people-first content and evaluating whether a site is genuinely satisfying users.
That means asking uncomfortable questions.
Is this genuinely one of the better pages available about the subject?
Does it add anything original?
Who wrote it?
Why should someone trust it?
Does the author have experience with the subject?
Is the information current?
Are we answering the query directly?
Is the page mostly written because a keyword had search volume?
Are ten pages saying essentially the same thing?
Could the content have been published by any competitor with the company name swapped out?
That last question is one I increasingly like.
If I can replace your brand name with your competitor’s and the article still makes perfect sense, what exactly makes the content yours?
Spam Update? Ask Different Questions
If the timing aligns with a spam update, my investigation changes.
Google says sites affected by spam updates should review its spam policies and that recovery can take time as automated systems reassess compliance.
So I’m looking much harder at how the website has been growing.
How was the content produced?
Are there hundreds or thousands of low-value pages?
Are we publishing at scale without adding much original value?
Are third parties publishing content primarily to take advantage of the domain’s authority?
Are there suspicious link-building patterns?
Have expired domains been used?
Is content scraped or lightly rewritten?
Are pages being created primarily to manipulate rankings?
A spam update isn’t the time to ask:
“How many times should we use the keyword?”
The question is whether the site’s SEO practices themselves are creating risk.
Sometimes the Correct Response Is No Response
This might be the hardest recommendation to give.
Imagine traffic drops 6% during an update.
Rankings move around.
A week after the rollout finishes, most of them recover.
What should you change?
Maybe nothing.
Not every fluctuation requires intervention.
Google specifically cautions against making drastic changes to content that’s already performing well after small ranking movements.
SEO can become dangerous when we believe every graph requires an action item.
Sometimes:
Monitor
is the action item.
Don’t “SEO” Your Way Out of an Update
This is another temptation.
Rankings fall.
So we add keywords.
Increase word count.
Add FAQs.
Rewrite every title.
Add schema.
Create 20 internal links.
Update publication dates.
And hope something sticks.
But if you don’t understand why visibility changed, you’re just creating more variables.
Worse, you can damage pages that weren’t actually broken.
I would rather make five changes based on evidence than 100 changes based on panic.
Document Everything
This is something I’ve become increasingly serious about.
When an update happens, record it.
Record when the rollout started.
Record when it ended.
Annotate your reporting.
Record major website changes around the same period.
Record which directories declined.
Record which page types improved.
Record which competitors gained.
Record what you changed afterwards.
Because six months later, another update will happen.
And suddenly that history becomes incredibly useful.
You can start asking better questions.
Did the same section decline again?
Did the pages we improved recover?
Are the same competitors gaining?
Did a content type that survived the previous update decline this time?
SEO gets much easier to analyse when you have a memory longer than your dashboard’s default date range.
Recovery Doesn’t Always Happen Immediately
This is another reason I wouldn’t promise a client:
“We’ll fix this and rankings will be back next week.”
Google says improvements can sometimes show within days, but it may take several months for systems to recognise longer-term changes. A site also doesn’t necessarily have to wait for the next major core update to see improvement.
That’s a much less exciting answer than:
“We found the issue!”
But it’s more responsible.
SEO recovery isn’t always a switch.
Sometimes it’s a process.
The Update Isn’t Your Strategy
This is probably the bigger lesson I’ve learned from watching Google updates over the years.
If your SEO strategy changes completely every time Google announces an update, you probably don’t have much of a strategy.
A good website should already be moving toward:
useful content;
original information;
strong technical foundations;
clear site architecture;
good internal linking;
real expertise;
credible brand signals;
good user experience;
accurate information;
and pages that genuinely deserve to rank.
Google updates may change how those things are evaluated.
But I don’t want to rebuild an SEO strategy around every piece of algorithm gossip on X.
I’d rather understand what changed, look at my own data, study the SERPs and make evidence-based decisions.
My Google Update Checklist Is Actually Pretty Boring
When Google announces another update—and it will—I don’t have a secret algorithm recovery playbook.
My process is much less dramatic:
- Identify the update
- Understand what type it is
- Record the dates
- Monitor during rollout
- Rule out technical and tracking problems
- Wait for the rollout to settle
- Compare before and after
- Identify the affected pages and queries
- Find patterns
- Study the winners
- Decide whether changes are actually necessary
- Monitor what happens next.
That’s it.
No panic.
No rewriting the entire website on day two.
No pretending I know exactly what Google changed when Google hasn’t told us.
And definitely no telling a client they’ve been “penalised” just because their rankings moved.
Google updates are part of SEO.
Some will hurt.
Some will help.
Some will pass without you noticing.
The skill isn’t predicting every update.
It’s knowing how to investigate what happened without making the situation worse.
And sometimes the smartest thing an SEO can do after a Google update is surprisingly difficult:
Wait until you actually have enough information to make a decision.
