Lets talk
Fill in the Form below for a no-string attached free technical discussion.
Google’s algorithm is the single biggest variable in organic visibility, and it has not stopped moving since the company released its first browser toolbar in 2000. The Google algorithm is a key factor in successful SEO optimization. Oftentimes a developer’s knowledge of the Google algorithm can make or break a good website. A developer or content team that does not track these updates is optimising for a version of Search that no longer exists.
This is why it is important to keep track of the latest updates and key feature changes in the algorithm. It is also important to know when these updates occurred and the influence they had on the SEO community.
This piece is the most complete version of that history we publish. It starts with the updates that built modern SEO and runs all the way through the AI-driven core updates of 2026, with every external reference linked directly to Google’s own documentation.
If you have used Moz’s algorithm change history page as your reference point in the past, you may easily relate with this list; a single, continuously maintained record you can check whenever a ranking drop sends you looking for a cause. We update this page every time Google confirms a new core, spam, or Discover update, so it stays current the way that kind of reference needs to.
Key Takeaways
- Google has confirmed many named ranking updates since the original 2000 Toolbar launch, and the Search Status Dashboard now logs every core, spam, and Discover update in real time since 2021.
- The single biggest shift since 2022 is the move from ranking pages to ranking answers; the Helpful Content Update, the March 2024 core update, and AI Overviews all reward original, people-first content and penalise pages built to match search queries rather than serve readers.
- AI Mode, launched in Search Labs in March 2025, passed 1 billion monthly users by May 2026 and now runs on Gemini 3.5 Flash globally. This considered Google’s clearest signal yet that conversational, AI-generated answers are becoming a permanent layer of Search, not an experiment.
- Core updates now arrive roughly every three to four months. The most recent, the May 2026 core update, rolled out from May 21 to June 2, 2026.
- The practical takeaway has not changed since 2022: content written by a named expert, with original data and clear E-E-A-T signals, is what survives every update on this list.
Google launched its browser toolbar, along with Toolbar PageRank, ensuring SEO arguments for years to come. This free browser plug-in let internet users search from any web page using Google’s quick, highly relevant search algorithm.
Toolbar PageRank, or TBPR, is essentially what created the need for SEO as a discipline. PageRank itself remains part of Google’s core ranking systems today, as detailed in Google’s official guide to Search ranking systems.
A large index extension, Latent Semantic Indexing (LSI), brought attention to anchor text relevance and prompted Google to implement link neighbourhoods. LSI improved Google’s capacity to understand synonyms and conduct more advanced keyword analysis.
Google’s current guidance still warns against the keyword-stuffing tactics this update began to penalise, detailed in Google’s spam policies for Google Search.
Google, Yahoo, and Microsoft introduced the no-follow attribute to combat spam and govern the quality of outbound links. No-follow helps remove unsafe links, such as spam blog comments. This was not a standard algorithm update, but it had a lasting impact on the link graph, one significant enough that Google revisited how it treats no-follow links 15 years later, as explained in Google’s official post on evolving nofollow.
Personalized search tapped directly into users’ search histories to automatically adjust results, rather than relying on manual settings or profiles as earlier personalisation attempts had. The initial impact was minor, but Google went on to use search history across nearly every other product it built. Google still documents how this works today in its guidance on managing personalized recommendations in Search.
Google combined its Maps data into the Local Business Center in March 2005, after creating the LBC and encouraging businesses to update their information on the platform. This move triggered several improvements in local SEO that are still visible in Google’s local ranking guidance today.
With Jagger, Google rolled out adjustments aimed at low-quality links such as reciprocal links, link farms, and paid links. Jagger rolled out in at least three stages between September and November 2005, with the largest impact in October, when low-quality and sponsored links suffered the heaviest losses.
This was not an algorithm change, but it altered the presentation of results permanently. Google combined traditional search results with News, Video, Images, Local, and other verticals into one results page. The old ten-blue-links SERP was effectively retired.
Caffeine was a new web indexing system that let Google crawl and store data far more efficiently. Google reported that the change expanded its index and delivered roughly 50% better results.
“Places” pages first appeared in September 2009 as part of Google Maps. Google Places formally rebranded the Local Business Center, integrated Places pages more closely with local search results, and added local advertising options.
Google Instant extended Google Suggest by displaying results as a query was typed. SEOs braced for major ranking disruption; the actual impact turned out to be minimal.
Google launched the +1 button next to results links in response to competition from Facebook and Twitter. Clicking +1 let users influence both organic and paid results within their social network.
Panda affected up to 12% of search results, a figure that came directly from Google. It penalised sites with thin content, content farms, and high ad-to-content ratios. Panda rolled out over months, reaching Europe as late as April 2011. Google has since folded the principles behind Panda into its guidelines on creating helpful, reliable, people-first content.
This addressed a real problem: low-quality “content farms” were outranking higher-quality, more relevant sites for the same queries.
Google, Yahoo, and Microsoft backed a unified structured data approach, generating the schema vocabulary still documented today in Google’s structured data guidelines. This pushed search results toward richer, more visual formats.
Google shifted Google+ social data and user profiles into search results, while adding a visible toggle to turn personalisation off. The update pushed SEO content to be more concise, but it was also widely seen as a vehicle for Google’s broader Google+ strategy.
After weeks of speculation about an over-optimisation penalty, Google released the Webspam Update, quickly nicknamed Penguin. It targeted keyword stuffing and other spam tactics and affected roughly 3.1% of English-language queries. Penguin’s spam-fighting role has since been absorbed into Google’s broader spam policies for Google Search.
Google expanded Knowledge Graph capability into Spanish, French, German, Portuguese, Japanese, Russian, and Italian. Google described this as more than a translation effort, but it extended entity recognition itself into new languages.
Google rolled out a deeper version of Penguin, internally significant enough that the team called it Penguin 2.0 rather than a minor refresh. Where the original 2012 release mainly checked a site’s homepage and top-level pages for spammy links, this version dug further into a site’s internal pages and affected roughly 2.3% of English-language queries, and had a meaningfully larger impact than most Penguin updates before or since.
This is also the update most often searched for under “google seo update april 2013,” since speculation about a major spring 2013 change had been building for weeks before Google confirmed the rollout in May.
Hummingbird was a full rewrite of Google’s core search algorithm, not a patch on top of it, the first since Caffeine in 2010. Google quietly switched search over to Hummingbird in mid-August 2013 and only confirmed it publicly on September 26, after testing showed no major disruption. Where Panda and Penguin filtered out specific categories of low-quality pages, Hummingbird changed something more fundamental: it let Google parse full, conversational questions and match intent rather than individual keywords, laying the technical groundwork for everything from voice search to the natural-language understanding Google later built on with RankBrain and BERT.
Google announced a targeted upgrade aimed at niches with a history of spammy results, naming payday lending and pornography specifically. Google said the rollout could take one to two months.
Pigeon significantly affected local results by changing how Google interpreted location signals, strengthening the connection between local and core ranking systems.
Following the removal of authorship photos on June 28, Google confirmed it would stop processing authorship markup entirely. Authorship bylines disappeared from search results by the next morning.
Google began penalising mobile pages that showed intrusive pop-ups or interstitials immediately on arrival, the kind that covered the main content and forced a visitor to dismiss an ad or app-install prompt before reading anything. Pages using less disruptive formats, like a legally required cookie notice or a reasonably sized banner, were unaffected. This was one of the earliest signals that user experience metrics would eventually become formal ranking factors, a thread that runs directly to the Page Experience Update covered later in this piece.
Google rolled out a significant, unconfirmed update that the SEO community nicknamed Fred after a passing joke from Google’s Gary Illyes. Fred targeted sites that prioritised aggressive ad placement and thin, low-value content over genuine usefulness to readers. Google never gave the update an official name or a public write-up, which is part of why “google algorithm update march 2017” and “fred update” both still generate steady search volume nearly a decade later.
Following criticism over offensive autocomplete suggestions and the prominence of fake news and misleading content in search results, Google announced Project Owl. The update improved Google’s ability to detect low-quality, deliberately misleading, or offensive content and gave users a direct way to flag problematic Featured Snippets and autocomplete predictions.
Google launched a jobs portal with a dedicated three-result block in search, pulling listings from sources including LinkedIn, Monster, Glassdoor, and CareerBuilder.
With Chrome 62, Google began warning visitors to sites with insecure forms. This was not an algorithm change, but it was a major step in Google’s push toward HTTPS and had a measurable effect on site traffic.
Google extended search snippet length across a wide number of results, effectively pushing meta description display length from roughly 155 to 300 characters. Google confirmed the change but did not detail the mechanics.
Videos moved from organic-style results with thumbnails into a dedicated video carousel, reshuffling previously tracked organic positions. The number of SERPs containing videos rose by roughly 60%.
Google confirmed that its BERT natural language processing model was rolling out in 70 languages worldwide. BERT improved Google’s ability to understand natural, conversational queries, and Google adjusted both its algorithm and underlying hardware to support it. Google’s own explainer on how AI is making information more useful covers the original BERT launch in detail.
Google decided that a Featured Snippet counts as a promoted organic result, so URLs shown in a Featured Snippet would no longer also appear as a separate organic listing. This had a direct effect on organic CTR and rank tracking.
Google launched what is more accurately described as passage ranking for US/English queries, the ability to rank a specific section of a page, not just the page as a whole. Google initially estimated the change would affect 7% of queries.
Google announced a change favouring detailed, first-hand reviews over short reviews and spammy affiliate content, initially in English only. Google has since published full guidance for creating helpful, reliable product reviews.
Google announced these as linked updates released back to back. The June update ran from June 2 to 12; the July update followed from July 1 to 12. As with all core updates, Google offered limited detail beyond confirming the rollout windows.
Google rolled out a global link spam fix over two to four weeks, targeting manipulative link patterns across multiple languages.
SEOs noticed a sharp increase in Google rewriting page titles in search results. Google confirmed the change in September and pulled back some of the adjustments after widespread complaints about result quality.
Google released a large-scale spam update in early November, followed by a separate core update on November 17. Timing of the update overlapped with Black Friday and prompted significant debate in the SEO community.
A refresh of the April 2021 update, again rewarding high-quality, first-hand product reviews. The rollout took roughly three weeks.
Google began rolling out the Page Experience Update, incorporating Core Web Vitals into ranking for both organic and Top Stories results. The rollout, delayed several times, ultimately ran through August 2021. Google’s Page Experience documentation still anchors the technical requirements behind this update today, and we cover the practical side of meeting those thresholds in our guide to Core Web Vitals and page speed optimisation.
Google released its first core update of 2022, confirming in the official Search Central announcement that it would take one to two weeks to fully roll out. Google reiterated its standing guidance; pages that lose visibility in a core update have not been penalised. Core updates reassess content quality across the entire web, and previously under-rewarded pages can also see gains.
This update expanded the product reviews policy beyond English-language content and added more detail on the first-hand testing and expertise signals Google looks for, building directly on the April and December 2021 versions.
This is the update that reshaped how every B2B content team has had to write since. Google introduced a dedicated classifier to demote content created primarily to perform well in search rather than to help an actual reader, and rewarded original, people-first content instead. Google’s guidance on creating helpful, reliable, people-first content became the reference document the entire SEO industry now writes against.
This was not a one-time event. Google released further helpful content updates through 2023, and by March 2024 folded the system directly into core ranking, a shift covered later in this piece.
A standard core update with the usual broad, unspecified effects across search rankings, separate from the spam update Google released later that same month. We cover both of these September and October 2022 updates together, including Google’s own SpamBrain explanation, in our dedicated Google Spam Algorithm Update post.
A link spam update specifically targeting manipulative link-building practices, distinct from the broader site reputation and scaled content abuse policies Google introduced in 2024.
Google closed out 2022 with two more updates in quick succession: a further link spam update, followed nine days later by the second Helpful Content Update of the year, continuing to refine the August classifier.
Another refinement of the product reviews policy, again emphasising demonstrable first-hand use of the product being reviewed.
A standard core update, notable mainly for its short rollout window relative to some of the longer updates that followed in 2023 and 2024.
Google released a core update and a further Helpful Content Update on the same day, continuing the pattern of pairing broad ranking changes with the ongoing helpful content classifier work.
A dedicated refresh of the helpful content classifier, separate from any core update, reinforcing that this system was now a standing part of Google’s ranking stack rather than a one-off launch.
Google released two more core updates in close succession through the autumn of 2023, with the November update taking close to 26 days to fully roll out — longer than most prior core updates.
This is the most consequential update on this list since Panda. In its official announcement, Google said it had been tuning ranking systems since 2022 to reduce unhelpful, unoriginal content, and that the March 2024 update folded that work directly into core ranking. Google estimated this update, combined with its prior efforts, would cut low-quality and unoriginal content in search results by 40%.
The update arrived with new spam policies targeting three specific abuse patterns: expired domain abuse, where old domains are repurposed to exploit their existing authority; scaled content abuse, where content is mass-produced with little regard for user value, including through unsupervised AI generation; and site reputation abuse, where third-party content is hosted on a trusted domain to exploit its ranking signals. Google gave sites a two-month window to address site reputation abuse before enforcement began. Site owners can review the current rules directly in Google’s spam policies for Google Search.
For any site that had scaled content production using AI tools without meaningful human oversight, March 2024 was the update that ended that strategy.
In the official Search Central post, Google said this update continued its work to surface content people find genuinely useful and reduce content made to perform well on Search rather than serve readers. Google also said the update was designed to better capture improvements sites had already made since March, including for small and independent publishers.
After roughly a year of testing as the Search Generative Experience, Google launched AI Overviews to all US search users at I/O 2024, with Google’s own announcement framing this as letting Search “do more than you ever imagined” through multi-step reasoning on complex queries. Google said hundreds of millions of US users would see AI Overviews within the week, with a goal of over one billion users by the end of 2024.
A wave of inaccurate AI Overview answers in late May 2024 forced Google to publicly walk back several edge cases within two weeks. By August 2024, AI Overviews had expanded to six more countries, including the UK, India, Japan, Indonesia, Mexico, and Brazil, and Google has continued expanding both query coverage and country availability through 2025 and 2026.
Google closed out 2024 with a core update in mid-November, followed by a spam update in mid-December, continuing the now-established cadence of separating broad core changes from targeted spam enforcement.
The first core update of 2025 ran from March 13 to March 27. As with every core update since 2022, Google’s standing advice was to keep producing original, people-first content rather than chase any specific technical fix. We covered this rollout as it happened, query by query, in our dedicated March 2025 Core Update post.
Google launched AI Mode as a Search Labs experiment for Google One AI Premium subscribers in the US, running on a custom build of Gemini 2.0. AI Mode left Search Labs in the US on May 20, 2025, at Google I/O, when Google removed the opt-in requirement and upgraded the underlying model to Gemini 2.5. By August 21, 2025, AI Mode had expanded to more than 180 countries and territories in English. Google’s I/O 2025 Search update credited AI Overviews with driving a measurable increase in Search usage for the query types where it appears.
This update ran from June 30 through July 17, 2025, fitting the now-typical three-to-four-month gap between core updates that Google has maintained since 2024.
Running from December 11 to December 29, 2025, this was described by several SEO trade publications as the most disruptive update of the year, ahead of the even larger volatility Google’s March 2026 update would bring.
Outside of ranking changes, Google added an AI-powered configuration feature to the Search Console Performance report, letting site owners describe the analysis they want in plain language rather than manually setting filters. Full details are in Google’s Search Central blog post on the feature.
This sat alongside a more consequential change six months later. Google’s launch of dedicated Search Generative AI performance reports on June 3, 2026, giving site owners their first direct view of how content appears inside AI Overviews and AI Mode. We break down what that data does and does not tell you in our piece on Google Search Console’s Gen AI performance reports.
Google released a broad update to the systems behind Discover, aiming to surface more locally relevant content, reduce sensational and clickbait material, and reward sites with demonstrated topic-level expertise; evaluated topic by topic rather than site-wide. Google’s official announcement used the example of a local news site with a dedicated gardening section earning gardening expertise, in contrast to a general movie review site that published a single gardening article.
A rapid spam update, completing in under 20 hours, refined Google’s SpamBrain systems to better detect manipulative practices including auto-generated content and cloaking.
Running for roughly 12 days through April 8, 2026, this was the most volatile core update site owners had seen in years, with a huge percentage of sites reportedly experiencing noticeable ranking changes. The update continued to weigh E-E-A-T and real-world credibility signals heavily, further reducing visibility for thin or generic AI-generated content while rewarding pages with demonstrable first-hand expertise.
Google published a formal guide, “Optimising your website for generative AI features on Google Search”, covering both AI Overviews and AI Mode for the first time. The guide explicitly retired AEO and GEO as separate disciplines, stating that optimising for generative AI search is simply optimising for the search experience still SEO. It also confirmed that Google’s AI uses a query fan-out technique, meaning a single well-scoped paragraph on a specific subtopic can earn an AI citation even when the page does not rank in the top ten for the broader query.
The same week, Google permanently removed FAQ rich results from Search, effective May 7, 2026, meaning FAQ schema no longer generates SERP features in any query type or language. We covered both changes in detail, including what each one means for B2B content teams, in Marketing Weekly Edition 002.
The second core update of 2026 ran from May 21 to June 2, shifting a huge number of top-three results across the queries tracked by major rank monitoring tools. This update was even more volatile than the March update that preceded it. Google offered no update-specific guidance beyond its standing direction. There is no targeted fix for a core update drop, and the most reliable path back is continuing to publish content that is genuinely useful to the people reading it. We tracked this update in real time, alongside the AI Mode and search box announcements landing the same week, in Marketing Weekly Edition 003.
At Google I/O 2026, Google confirmed that AI Overviews had reached 2.5 billion monthly users and AI Mode had passed 1 billion monthly users, a figure that quadrupled between May and November 2025, then doubled again over the following six months. Google announced Gemini 3.5 Flash as the new default model for AI Mode globally, alongside the biggest redesign of the search box itself in over two decades and the introduction of persistent “information agents” that monitor a topic continuously and notify users when something relevant changes. Full details are in Google’s official I/O 2026 Search announcement.
For content teams, this is the clearest signal yet that AI-generated, conversational answers are not a temporary layer sitting on top of classic Search; they are becoming the primary interface for an increasing share of queries.
Every update on this list, going back to Panda in 2011, rewards the same underlying thing: content written by someone who has actually done the work, for a reader who has a real question. The mechanism has changed; link quality, then content depth, then helpfulness classifiers, then AI-generated answers- but target and best practices are still the same.
If your content strategy still treats “what does Google want” as a moving target, the gap between your site and a competitor’s will keep widening with every update on this page. We’ve broken down exactly how to build content that survives core updates and earns AI citations in our guide to E-E-A-T signals that actually move rankings, it picks up directly where this history leaves off.
Entity signals are an increasingly important part of that equation. If Google’s Knowledge Graph cannot confirm your brand as a distinct entity, no amount of content depth will earn you a citation in AI Overviews or AI Mode, which is why we cover the implementation details in our piece on Google’s Organization schema data update.
Saurabh Garg, the visionary Chief Technology Officer at Whitebunnie, is the driving force behind our cutting-edge innovations. With his profound expertise and relentless pursuit of excellence, he propels our company into the future, setting new standards in the digital realm.
The Confirmation B2B Marketers Have Been Waiting For For years, Google gave a consistent answer whenever anyone asked whether social […]
Read Post »
Google has announced that the September 2022 core algorithm upgrade is now complete. The September 2022 core update was launched […]
Read Post »
Google has officially rolled out its March 2025 Core Update, marking its first significant algorithm update of the year. This […]
Read Post »
Copyright © 2026 White Bunnie -All Rights Reserved
