Google Core Updates 2024 to 2026: The Pattern We Found Across 43 Client Sites After Analysing 18 Months of GSC Data

Google announces a core update every quarter and we never know if we are safe or vulnerable until it hits. Three consecutive updates have either hit us or missed us by coincidence and I cannot see a pattern.” RankFast tracked 43 client sites through three Google core algorithm update 2026 cycles to answer exactly that question, and the pattern was not where most sites are looking.

It was not content word count. It was not link count. It was not domain authority, and it was not technical performance. It was the proportion of a site’s key topical pages attributable to a named, verifiable expert, the kind of signal that shows up nowhere in a standard technical audit.

Here is what the update cycle actually targeted, the pattern that separated winners from losers, a real caveat about how much that pattern actually proves, and a scoring framework to check your own site’s vulnerability before the next one lands.

Key Takeaways

• Across 43 client sites tracked through three core updates (August 2024, March 2025, and December 2025), the strongest predictor of positive versus negative impact was the proportion of key pages with a named, verifiable expert author, not word count, links, domain authority, or technical performance.
• Sites with 40 percent or more of key topical pages carrying this signal maintained or improved rankings across all three updates. Sites below 20 percent declined in at least two of the three, according to RankFast’s tracked data.
• Google’s Search Quality Rater Guidelines list Trust as the foundational, most important element of E-E-A-T, and are explicit that E-E-A-T is not a direct ranking factor; it is a framework human raters use, and rater feedback trains the ranking systems rather than scoring each page live.
• Author signal correlating with survival does not prove it caused the survival. Sites resourced enough to credential expert authors likely have other advantages too, which is a real limitation of this data worth stating plainly rather than glossing over.

What The 2024 To 2026 Core Update Cycle Actually Targeted

Across 43 client sites tracked through three major core update periods, August 2024, March 2025, and December 2025, the variable with the strongest predictive power for positive versus negative impact was not content word count, link count, domain authority, or technical performance. It was the proportion of indexed pages with content attributable to a named, verifiable expert in the site’s primary topic area. While a technical SEO agency can identify issues with crawling, indexing, site structure, and page performance, the data suggests that technical health alone did not explain which sites maintained visibility across the three update periods. Sites where 40 percent or more of key topical pages had a named author with a verifiable sameAs link to a professional profile maintained or improved rankings across all three google algorithm update 2024 2025 periods tracked. 

Update	Confirmed Date	Type
Core Update	August 15 to September 3, 2024	Broad core update
Core Update	March 13 to 27, 2025	Broad core update
Core Update	December 11 to 29, 2025	Broad core update

Dates confirmed against Google’s Search Central blog and Search Engine Land’s update tracking. Note that June 30 to July 17, 2025 was also a confirmed core update inside this window, not included in RankFast’s three-update comparison set.

The Pattern That Separated Winners From Losers

Three characteristics were shared by the 18 sites that improved rankings across all three tracked updates, out of RankFast’s 43-site set. A bylined expert author appeared on every piece in the site’s primary YMYL or expertise-required category, with a linked bio and a sameAs credential link. Each piece carried at least one original data point or first-hand observation not available by combining other published sources. Content carried last-reviewed or last-updated dates showing a revisit within the past 12 months, with the review date visible in both the page HTML and the schema markup. Not every winning site had all three core update winners losers characteristics perfectly executed, but every one had at least two applied consistently across its primary traffic-generating pages.

Is This Really About Authors, Or Something Correlated With Having Them

A real limitation worth stating plainly: correlation between author signal and ranking survival does not prove authorship caused the survival. Sites with the budget and editorial process to credential named expert authors on every piece likely also have other advantages, larger content teams, more established brand history, and better existing trust signals, that could be doing some or all of the actual work. Author bylines could be a visible symptom of a well-resourced, already-trustworthy operation rather than the specific lever Google’s systems reward.

Google’s own Search Quality Rater Guidelines add useful context here rather than settling the question. E-E-A-T, Experience, Expertise, Authoritativeness, and Trust, is explicitly not a direct ranking factor computed live for each page. It is a framework human quality raters use to evaluate search results, and Trust is described as the foundational, most important element: a page cannot be considered high quality regardless of expertise if it fails on trust (Google Search Quality Rater Guidelines). Rater feedback trains and calibrates Google’s ranking systems over time rather than scoring an individual page directly. That is consistent with RankFast’s data (an algorithm that has learned to approximate what raters reward would plausibly favour verifiable authorship) without proving author signal specifically, rather than the trust it often correlates with, is the mechanism doing the work.

A cleaner way to test the confound would be comparing two sites with genuinely similar resourcing, similar age, similar existing backlink profiles, where only one added credentialed authorship. RankFast’s 43-site set was not built as that kind of matched comparison; it was an observational sample of existing clients, which is a real constraint on how strong a causal claim this data can support. That does not make the pattern useless. It means the honest framing is ‘sites with this signal tend to do better, and here is a plausible mechanism why,’ not ‘add author bylines and you are protected,’ and the recommendations later in this piece are worth doing on their own merits, not solely because they moved a metric in one dataset.

How To Assess Your Site’s Vulnerability Before The Next Core Update

Learning how to survive core updates starts with scoring, not guessing. Score your 20 highest-traffic organic pages on three criteria: does each page have a named author with a linked bio and a professional credential, does each page contain at least one piece of original information not present in the top 5 results for its target query, and is each page last-reviewed within the past 12 months. An organic SEO agency can use the same type of review to identify which pages rely on generic information, lack clear expertise, or have not been meaningfully updated. 

Score	Risk Level
0 of 3 criteria met	High risk
1 of 3 criteria met	Medium risk
2 to 3 of 3 criteria met	Low risk

If 30 percent or more of your 20 highest-traffic pages score 0 of 3, the site has measurable core update vulnerability by this framework.

What Changes Actually Prevented Core Update Impact

In the 23 sites that maintained stable rankings across all three update periods, three changes were implemented before the first update hit, forming the basis of RankFast’s core update recovery SEO approach for existing clients. A named, credentialed author with a sameAs link to a verifiable profile was added to every piece in the primary topical area. A last-reviewed date was added to every piece that had not been edited in the past 12 months, paired with an actual factual accuracy review, not just a date change. Pages under 200 words of original content with no first-hand information were removed or noindexed entirely. These are the kinds of foundational improvements that should form part of an ongoing SEO service, rather than being treated as emergency work after a ranking decline. 

These three address the author trust signal, the content freshness signal, and the content quality dilution signal simultaneously, rather than treating any one of them as sufficient alone. Sites that implemented all three were the most protected in RankFast’s tracked set.

None of the three requires new technology or a platform migration. All three are editorial and content-operations work: assigning real bylines, actually reviewing content rather than just updating a timestamp, and being willing to remove pages that were never going to perform. That is a less exciting answer than a technical fix, which may be part of why it gets less attention relative to how consistently it showed up across this dataset.

What This Means For Your Google Quality Update Strategy In 2026

A google quality update SEO strategy built around this pattern treats every core update the same way regardless of what SEO commentary speculates the update specifically targeted. Chasing the specific rumoured focus of each individual update is less reliable than maintaining the same three structural signals continuously: verifiable authorship, genuine content freshness, and removing thin pages that dilute a site’s overall quality signal. An AI SEO agency may also examine how these signals affect the way content is understood across increasingly sophisticated search and AI-assisted discovery systems, but the underlying principle remains the same: build content around genuine expertise, accuracy, and useful information. 

This also reframes what a core update recovery plan should look like after a site is already hit. The instinct is usually to look for what changed specifically in that update’s announcement and reverse-engineer a fix targeted at it. RankFast’s data suggests a more durable approach: audit the same three signals, authorship, freshness, and content depth, across the whole site rather than the pages that dropped specifically, since the next update is unlikely to target the exact same narrow slice twice. A site that fixes only the pages that fell in one update is optimising for the last update, not the pattern behind all of them.

Score Your Own Site This Week

This uses the same scoring framework covered above, applied directly to your own analytics. While the framework here focuses on organic search visibility, businesses often combine SEO with search engine marketing services to maintain visibility across both organic and paid search while longer-term SEO improvements are being implemented. 

1. Pull your 20 highest-traffic organic pages from Google Search Console, sorted by clicks over the last 12 months.
2. Score each one on the three criteria: named credentialed author, original information not in the top 5 competing results, and a genuine review within the past 12 months.
3. Calculate what percentage scored 0 of 3. If it is 30 percent or higher, treat the site as having measurable vulnerability.
4. Prioritise fixing the highest-traffic 0-of-3 pages first, since these carry the most exposure if the pattern holds for the next update.

The Pattern Is Real. The Certainty Should Not Be.

Forty-three sites showing the same pattern across three real, confirmed core updates is meaningful evidence. It is not proof that author signal specifically, rather than the trust and resourcing it tends to travel with, is the exact lever Google’s systems reward. Treating this pattern as a checklist to complete, rather than a certainty to bank on, is the more honest way to use it, and it is also the more useful one. Verifiable authorship, genuine freshness, and removing thin content are worth doing regardless of which specific mechanism explains why they correlate with survival. An experienced SEO agency can use this framework as part of a broader site audit rather than treating any single ranking signal as a guaranteed solution. 

Frequently Asked Questions