Google Algorithm Update 2026: 5 Rollouts, 80% Volatility

By Alex Bain 2026-07-06 15 min read

Google confirmed five separate ranking updates in the first six months of 2026, and the March core update alone reshuffled 79.5% of all top-three positions, the most volatile broad rollout since December 2025. The pattern across every update is consistent: Google tilted visibility toward first-party, brand-owned, and government sources while stripping traffic from AI content farms, comparison aggregators, and pages built primarily to rank. For B2B marketing and SEO teams, the google algorithm update 2026 cycle is not a single event to recover from. It is a sustained reweighting of what counts as trustworthy, and the sites that treated it as a one-time storm are the ones still down today.

The 2026 Google algorithm update timeline at a glance

Google shipped updates at a faster cadence in 2026 than in any prior year on record. Where the company historically ran two or three broad core updates annually, the first half of 2026 delivered two confirmed core updates, one Discover-specific core update, and two spam updates, plus one unconfirmed volatility event that Google never claimed. Core updates now land roughly every three months rather than twice a year. That compression matters because it shortens the window teams have to diagnose damage, ship fixes, and wait out the mandatory measurement period before the next wave arrives. The table below lists every confirmed 2026 update with its documented rollout window and primary effect, drawn from Google's Search Status Dashboard and reporting by Search Engine Land.

The cadence tells the story. Google ran the fastest spam update in its history on March 24, wrapping in under a day, then opened a 12-day core update three days later. Six weeks after that closed, the May core update ran nearly 12 days more. Add the two-day June spam update and a publisher spent most of the second quarter of 2026 with at least one algorithm in flight. The February Discover update is easy to overlook because it touched the Discover feed rather than classic search, but it set the tone for the year: reduce sensational content and clickbait, and surface in-depth, original, and timely work from sites with genuine expertise in a topic.

Inside the March 2026 core update, the most volatile rollout in years

The March 2026 core update opened on March 27 and finished on April 8, and by every third-party measure it was the most turbulent broad update in more than a year. SE Ranking's dataset put top-three volatility at 79.5%, up 12.7 percentage points from the 66.8% recorded during the December 2025 core update. Top-ten volatility climbed to 90.7% and top-100 volatility reached 98.5%. SEMrush Sensor peaked at 9.5 out of 10 during the rollout, among the highest readings the tool has ever logged. Nearly one in four pages that started in the top ten fell out of the top 100 entirely.

The most revealing figure is the last one. In December, 13% of top-three results were pages that had been sitting outside the top 20. In March, that jumped to 29.7%. Google was not nudging the existing order. It was pulling in pages that had been buried and demoting incumbents wholesale. SE Ranking also found that domain age barely moved: sites 15 years or older still held roughly 57% of visible positions, with a median domain age between 17.3 and 17.5 years. Established domains kept their structural advantage, but individual pages within them were still exposed if the content did not meet the bar. Our SEO team reads that combination as a durability signal, not an age pass: legacy authority buys resilience, not immunity.

March 2026 winners and losers, from YouTube to GoodRx

Lily Ray, Vice President of SEO and AI Search at Amsive, framed the March update as a clean break from the pattern that came before it.

"If December 2025 read primarily as an eCommerce reset, March 2026 looks like something different: a first-party, official-source correction." Lily Ray, Amsive.

The visibility data backs the framing. YouTube was the single largest loser in absolute terms, shedding 567 visibility points, a 13.4% decline. User-generated and aggregator platforms followed it down: Reddit, Instagram, TripAdvisor, Indeed, ZipRecruiter, and Glassdoor all lost ground. Government domains, official brand sites, and category-defining first parties gained. The table below samples the swing.

The winners share a trait. They are the primary source for the thing being searched. IMDb owns film metadata, Spotify owns the track, GoodRx owns drug pricing, and the .gov domains own the official answer on student aid and federal jobs. Aggregators that repackaged those answers lost. Third-party trackers reported that affiliate sites saw drops at a 71% rate and that AI content farms lost 60% to 80% of their traffic during the same window. Google did not announce an AI penalty, and its public position remains that it rewards helpful content regardless of how it is produced. But the measured outcome was brutal for thin, scaled, and derivative pages. Teams that depend on aggregation or affiliate models should read the full Amsive winners and losers analysis before their next content sprint.

The May 2026 core update hit harder and reached deep into YMYL

If March was the most volatile, May was the most powerful. The May 2026 core update began on May 21 at 8:40 a.m. PDT and completed on June 2 at 5:40 a.m. PDT, a rollout of 11 days and 21 hours. SEMrush Sensor peaked at 7.0 out of 10, classified as high, with multiple spikes across the twelve-day window. Glenn Gabe of G-Squared Interactive tracked it hour by hour and was blunt about its force.

"The May 2026 core update just completed. It was a big one, more like a typical core update." Glenn Gabe, G-Squared Interactive.

Gabe's day-by-day notes captured how fast it moved. He wrote that the update 'looks like it is landing already' within 48 hours, and summed up the two rollouts side by side as 'March was meh, but May is big.' The categories that moved hardest were Your Money or Your Life verticals. Gabe flagged serious volatility in gambling, which he described as an ultra-YMYL category that can swing violently during major updates, along with early and heavy movement in finance and health. News and Top Stories features also reshuffled. One notable case: Grokipedia surged during May after crashing in the February Discover update, a reminder that a single site can whipsaw across consecutive rollouts.

For operators in regulated or high-stakes niches, the May update reinforced a hard rule. YMYL topics carry the tightest trust requirements, and Google will move them the most when it recalibrates. Sites covering medical, financial, legal, or gambling content without verifiable authorship, credentials, and citations were the most exposed. Read Gabe's full May 2026 case analysis for the specific site trajectories he documented.

Why 2026 core updates take almost two weeks to roll out

A recurring question in every 2026 update thread was why a core update needs 12 days to finish when a spam update can wrap in 20 hours. John Mueller of Google addressed the mechanics directly, and his explanation reframes what a core update actually is.

"We generally don't announce 'stages' of core updates. Since these are significant, broad changes to our search algorithms and systems, sometimes they have to work step-by-step, rather than all at one time. It's also why they can take a while to be fully live." John Mueller, Google.

Mueller added that there is no single core update machine that gets switched on with an identical flow each time. Instead, Google ships changes based on what individual teams have been working on, and those systems and components change from one update to the next. That detail explains the multiple tremors practitioners logged during both the March and May rollouts. What looks like one algorithm shift is actually several ranking components, tied to different engineering teams, deploying in sequence across a two-week window. A site can drop on day two, appear to recover on day six, then drop again on day ten as a later component lands.

The practical consequence is measurement discipline. Google's own core update documentation tells site owners to wait at least a full week after a rollout completes before drawing conclusions, then to compare that week against the week before the update began. For the May update, that put the earliest clean comparison window at roughly June 9. Teams that panicked and rewrote pages mid-rollout often misread a temporary tremor as a permanent loss. PPC Land's breakdown of Mueller's rollout mechanics lays out the full exchange.

The spam updates: March's record sprint and June's SpamBrain tune-up

The two 2026 spam updates operated on a different clock and a different logic than the core updates. The March 2026 spam update began and finished on March 24, wrapping in just under 20 hours, the fastest spam rollout Google has ever run. It targeted automated and scaled spam, and it landed three days before the March core update opened. That sequencing produced a compounding effect that SE Ranking quantified: 82% of domains that dropped during the spam update stayed gone through the core update, and only 23.5% of domains that lost top-ten positions clawed back into the top ten. Getting hit by the spam update was, for most sites, a one-way door.

The June 2026 spam update ran from June 24 at 9:00 a.m. PT through June 26, roughly two days, and rolled out globally across all languages. Google described it as improvements to automated spam detection, including SpamBrain, its machine-learning spam-prevention system, and notably did not announce any new spam policies. That is an important distinction. A spam update enhances how existing rules are enforced. A core update changes how quality itself is weighed. Sites that confused the two spent June chasing the wrong fix.

Spam updates also differ in reversibility. Google's guidance is that recovering from a spam action requires bringing a site back into policy compliance and can take months, whereas core update recovery hinges on demonstrable content quality improvements assessed at the next broad update. For the specifics, Search Engine Land covered the June 2026 spam update rollout as it happened. Teams with any history of scaled content generation should treat the spam-then-core sequence as the template for the rest of 2026.

What the 2026 updates actually reward

Strip away the volatility charts and a single consistent profile emerges from all five 2026 updates. Google is rewarding the primary source and punishing the repackager. Across March and May, the sites that gained were the original owners of the information being searched, the domains with verifiable expertise, and the pages a human clearly reviewed. The sites that lost were aggregators, thin affiliates, and scaled or templated pages with no demonstrable author, experience, or first-hand data. The following signals separated winners from losers in 2026, based on the March and May outcome data:

• First-party ownership of the answer. IMDb for film, GoodRx for drug pricing, and .gov domains for official processes all gained because they are the source, not a summary of it.
• Verifiable authorship and credentials. Content with a named, qualified author and transparent attribution held up in YMYL verticals where anonymous pages collapsed.
• Original data and first-hand experience. Pages citing proprietary research, case studies, or lived experience outperformed pages that rephrased existing material.
• Topical depth over topical breadth. Sites with genuine focus in one area beat generalist domains publishing shallow coverage across many.
• Technical health. Core Web Vitals and interaction metrics such as INP remained a tiebreaker, especially where content quality was otherwise comparable.
• Editorial oversight. A human review layer, not just a publish button, correlated with resilience through every rollout.
• Restraint on scaled output. Sites that slowed AI-assisted publishing and cut thin pages generally fared better than those that flooded the index.

None of these are new in principle. E-E-A-T, which stands for Experience, Expertise, Authoritativeness, and Trustworthiness, has been in Google's guidelines for years. What changed in 2026 is enforcement intensity. The gap between sites that merely claimed expertise and sites that could prove it widened at every update. If your content strategy cannot survive the question of who wrote this and why anyone should trust them, 2026 was the year that question started costing rankings. Structuring content for both classic search and AI answer engines is where our answer engine optimization work concentrates.

The zero-click squeeze and the new traffic economics

The 2026 updates did not happen in a vacuum. They landed while the underlying economics of search traffic shifted against publishers. SparkToro's 2026 study, built on Similarweb's desktop and mobile panel for January through April 2026, found that 68.01% of Google searches in the United States ended without a click to any website, up from 60.45% in 2024. That 7.56-point jump in two years is the fastest acceleration SparkToro has recorded since it began tracking the metric. The share of searches that led only to another Google search rose 7.2 points, as AI Overviews and direct answers absorbed intent that used to become a visit.

Set that against the money. Alphabet's fiscal 2025 10-K reported Google Search and other revenue of $224.53 billion, more than half of Alphabet's total, and the company crossed $400 billion in annual revenue for the first time. Search is not shrinking as a business. It is growing while sending a smaller share of clicks outward. For B2B teams, that is the core tension of 2026: the algorithm updates raised the quality bar to rank at the same time that ranking returns fewer clicks per query.

The strategic response is not to abandon SEO but to change what a ranking is worth. A top-three position that also earns a citation inside an AI Overview compounds visibility even when the click does not come. That is why our generative engine optimization and technical SEO work now treat AI-answer citations as a first-class ranking outcome, not a side effect. SparkToro's zero-click findings and Search Engine Land's coverage of the study hold the full dataset. With Google holding roughly 90% of global search share per StatCounter, there is no volume alternative large enough to route around.

How 2026 compares to Panda, Helpful Content, and prior years

The 2026 cycle rhymes with a decade of Google quality updates, but the tempo and the target changed. When Google launched Panda in 2011, it ran quality refreshes intermittently and often with long gaps. The Helpful Content System, introduced in August 2022, was Google's first attempt at a site-wide classifier aimed at content written for search engines rather than people. In March 2024, Google folded the helpful content signals into its core ranking systems, ending the separate helpful content update and making that judgment a permanent part of every core update. The 2026 updates are the mature form of that 2024 decision. There is no separate helpful content switch anymore because helpfulness is now assessed inside the core.

Two things distinguish 2026 from the 2023 and 2024 cycles. First, cadence. Google ran core updates roughly every three months in the first half of 2026 rather than the historical two to three per year, which means less recovery runway between rollouts. Second, the AI content dimension. The 2022 and 2023 updates predated the flood of generative content. By 2026, entire site categories had been built on scaled AI output, and the March and May updates functioned as the correction. The 60% to 80% traffic loss that AI content farms absorbed in March has no clean analog in the Panda era because the volume of low-effort content is now orders of magnitude larger.

The through line from Panda to 2026 is unchanged: Google periodically re-decides what quality means and redistributes traffic accordingly. What is different is that the redistribution is faster, deeper into YMYL, and aimed squarely at content that no human meaningfully touched. Google's February 2026 Discover announcement used the same language it has used for years, in-depth, original, and timely content from sites with expertise, which is the real tell. The words did not change. The enforcement did.

Predictions for the rest of 2026 and into 2027

Based on the trajectory of the five confirmed updates and Google's public signals, here is where the google algorithm update 2026 cycle most likely heads next. These are directional calls, not certainties, and each carries a concrete implication for how B2B teams should allocate effort in the next two quarters.

• Expect a third 2026 core update in the September to October window. The roughly three-month cadence from March to May points to another broad update before Q4, likely with continued YMYL emphasis.
• The February Discover update will expand beyond US English. Google said it would reach more countries and languages in the coming months, so international publishers should prepare for Discover volatility in the second half of 2026.
• AI Overview citations become the metric that matters. As zero-click passes 70% in some verticals, teams will increasingly measure success by inclusion in AI answers, not just blue-link position.
• Spam updates keep getting faster and more frequent. The record 20-hour March rollout and the two-day June rollout suggest SpamBrain iterations will land with little warning, making continuous compliance more important than post-hoc cleanup.
• Aggregator and affiliate models face further compression. The first-party correction that defined March is a direction, not a one-off, and thin comparison content will keep losing ground to primary sources through 2027.

The teams that will win the next 18 months are the ones treating these updates as a standing operating condition rather than a series of emergencies. That means building a content and authority engine that assumes quarterly recalibration. Durable off-site authority is part of that engine, which is why disciplined link building remains a hedge against volatility: the domains with the deepest earned authority absorbed the 2026 shocks best, exactly as the domain-age data predicted.

What to do Monday morning

If your traffic moved during any 2026 update, the worst response is a frantic rewrite mid-rollout. Google's guidance and the 2026 data both argue for disciplined diagnosis first. Here is the sequence to run this week, in order.

1. Confirm the timeline. Map your traffic changes against the exact update windows: February 5 to 27, March 24, March 27 to April 8, May 21 to June 2, and June 24 to 26. A drop that predates every window is not an algorithm problem.
2. Separate spam from core. If you fell during a spam update and stayed down, the fix is policy compliance, not content polish. The SE Ranking data showed 82% of spam-update casualties never recovered, so treat this as urgent.
3. Wait the full week. Do not draw conclusions until seven days after a rollout completes, then compare that week to the week before it began. Mid-rollout tremors are not verdicts.
4. Audit for first-party value. For every important page, answer whether you are the source or a summary. Summaries lost in 2026. Add original data, named expert authorship, and first-hand experience where you can.
5. Fix the technical tiebreaker. Bring Core Web Vitals and INP into the green. When content quality is close, technical health decided the position in 2026.
6. Cut, do not just add. Prune thin, scaled, and derivative pages. Sites that reduced low-value output generally recovered faster than sites that only published more.
7. Instrument AI visibility. Track whether you appear in AI Overviews for your priority queries, because a citation without a click is now a real outcome worth defending.

None of this is quick, and that is the point. The 2026 updates rewarded organizations that treat search as a durable capability rather than a campaign. If you want an outside read on where you stand, a structured AI visibility audit against the 2026 signal set is the fastest way to separate what actually moved your rankings from what merely felt urgent. Start with the diagnosis, ship the highest-value fixes first, and measure at the next update. That is the whole 2026 playbook.

Alex Bain

Head of Growth

Alex leads programmatic SEO and AEO at Skitrate. Twelve years in growth, previously head of organic at a B2B SaaS unicorn and an ex-agency SEO director. Specializes in scaling content from zero to seven-figure traffic on technical and high-intent commercial verticals.

• 12 years leading SEO and growth at agencies and in-house SaaS
• Built and scaled programmatic SEO catalogs across SaaS, fintech, edtech
• Active practitioner of AEO/AIO/GEO measurement and citation tracking
• Public talks at SearchLove, BrightonSEO, MozCon

More posts by Alex Bain →