DemandSphere last week published a free tool that maps every confirmed Google Search ranking update and AI search milestone since 2000 - a resource the company says it built because nothing comparable existed. The Google Algorithm & AI Search Update Tracker, released on April 18, 2026, covers 173 entries spanning 25 years, combining live data from Google's own status dashboard with DemandSphere's internal annotation database and Google's published research papers.

Ray Grieselhuber, founder and CEO of DemandSphere, announced the launch on LinkedIn, describing it as "something I've wanted for a long time on our site." The post drew immediate engagement from SEO professionals, with Tim Resnik calling it a tool he added "to my stack of SEO resources to regularly check," and Joshua Squires describing it as "pretty sick."

The tracker is available at demandsphere.com/research/demandsphere-radar/algorithm-update-tracker, with a full dataset accessible via a JSON API at the /api.json endpoint. The data is published under CC BY-NC 4.0, requiring attribution but carrying no authentication requirements and no rate limits.

What the tracker contains

The database catalogues 173 updates tracked between 2000 and 2026. According to DemandSphere, there are 52 core updates - broad ranking system overhauls affecting all content types, averaging 17 days each - and 26 spam updates, which target manipulative tactics and can complete in under 24 hours. The longest rollout recorded in the dataset is the March 2024 core update at 45 days. The most recent entry is the March 2026 core update, which ran from March 27 to April 8, 2026 - a 12-day cycle that PPC Land covered in detail as it rolled out.

Eight major AI milestones are tracked separately within the tool, flagged with a distinct icon to differentiate them from ranking enforcement events. These milestones run from Hummingbird in 2013 through AI Mode in 2025. Each milestone entry links to the relevant Google research publication or product announcement.

Each record in the tracker includes a date, a type classification (Core, Spam, AI, System, or Other), a duration where known, and a researched description of what changed and what was affected. Every entry has a hash link - so a URL like #bert or #florida loads directly to that row and expands its detail panel. The tracker includes filtering by type and year, name-based search, sortable columns, and a frequency chart that breaks update volume down by year with AI milestones annotated above the bars.

Three data sources

DemandSphere draws on three distinct inputs. The first is Google's Search Status Dashboard at status.search.google.com, fetched live on every page load. This source covers the period from 2021 to the present, which means any update Google confirms officially now appears in the tracker automatically.

The second source is DemandSphere's own historical annotation database, which the company says has tracked Google algorithm changes for more than 20 years. This covers the pre-Dashboard era: the Florida update of 2003, Panda, Penguin in 2012, Hummingbird, and every named update through 2020. According to DemandSphere, this data was imported directly from its production systems rather than reconstructed from secondary sources.

The third source is Google's research publications. AI milestones such as the 2017 Transformers paper and the 2019 BERT deployment are sourced from Google's own papers and product announcements rather than from community blog posts or industry trackers, which DemandSphere notes have historically been scattered and incomplete.

The argument that Google search has been AI-powered for over a decade

The release carries a thesis. According to DemandSphere, the popular framing that AI search began in 2023 with ChatGPT or the Search Generative Experience is incorrect. The company argues that Google has been weaving machine learning into its ranking systems since 2013. "Google has been weaving AI into its ranking systems since 2013," the blog post states. "This is why we always say 'it's ALL AI search.'"

The case rests on a specific sequence of milestones. In 2013, Hummingbird rewrote the core algorithm for semantic understanding, shifting query processing away from keyword matching toward intent interpretation. Two years later, RankBrain introduced machine learning to handle the roughly 15% of queries Google had never encountered before. Then, in 2017, Google Research published "Attention Is All You Need" - the paper that established the Transformer architecture now underlying GPT, BERT, Gemini, and most major language models in deployment today.

BERT arrived in 2019, applying deep natural language processing to 100% of English queries. Google described it at the time as the biggest leap in search quality in five years. In 2021, MUM - described by DemandSphere as "1,000x more powerful than BERT" - added multimodal understanding across text, images, and 75 languages. AI Overviews followed in May 2024, placing generative AI output directly into the main results layout. PPC Land documented the launch and its implications for click-through rates as the feature rolled out. AI Mode, which DemandSphere classifies as fully conversational search with a chat-style interface and cited web sources, arrived in March 2025.

Google's head of search, Liz Reid, acknowledged the long arc of this development in October 2025: "I think it's been an evolution in Search over the last several years, but it's exciting to be able to bring it more forefront to the user and unlock new capabilities that way," as reported by PPC Land. BERT, Reid noted, "was incorporated into search results before ChatGPT existed."

The consequence, according to DemandSphere, is that organizations focusing exclusively on AI Overviews when assessing the impact of AI on search "are looking at the last few steps of a long-running project."

Why the tool was built

DemandSphere monitors brand visibility across Google SERPs, AI Overviews, ChatGPT, Perplexity, and other AI search platforms. The practical trigger for building the tracker, according to the company, was a recurring question from customers: when a Google core update rolls out, they want to know immediately whether they are seeing an algorithm change or something else.

Google's Search Status Dashboard only goes back to 2021. The SEO community's historical records - blog posts, archived tweets, third-party tools - are fragmented and do not place algorithm updates alongside AI research milestones on the same timeline. DemandSphere describes building its own dataset in its Annotations database specifically so customers can overlay any of these changes on performance charts within its dashboards.

Grieselhuber's LinkedIn post framed this as a shared problem: "We built this primarily because it's a tool we needed ourselves so we can integrate it into our platform but, as always, if one team needs it, usually that means many more do as well."

The March 2024 core update is the clearest illustration of why timeline context matters. That rollout, which DemandSphere's tracker records as running 45 days - the longest in the dataset - absorbed the helpful content systemdirectly into core ranking. It was preceded by a spam update that launched on the same date, March 5, 2024, creating the same attribution difficulty that recurred when Google released the March 2026 spam and core updates within 72 hours of each other. PPC Land tracked the 2026 pair as they landed, noting that isolating the cause of any ranking shift during overlapping rollouts is "nearly impossible."

How SEO professionals can use it

The tracker is not designed as a retrospective archive only. According to DemandSphere, when ranking changes appear in Google Search Console, practitioners can open the tracker and check the timeline to determine whether a confirmed update was rolling out during their volatility window. If a confirmed update is active, the tracker shows its type, known duration averages, and historical context for what that type of update typically targets.

The type filters enable a different kind of analysis. Looking at all core updates across 25 years alongside all spam updates reveals how enforcement has evolved - from the early Penguin-era link scheme targeting to SpamBrain's AI-driven detection. The frequency chart makes the acceleration of AI milestones visible: the gap between each milestone has shortened substantially since 2021.

For developers and data teams, the JSON API provides programmatic access to all 173 entries with no authentication required. According to DemandSphere, the API can be pulled by date range, filtered by update type, and fed into reporting systems or alerting workflows. The CC BY-NC 4.0 license permits non-commercial use and modification, with attribution required.

Context for marketing professionals

The release sits inside a period of significant algorithmic activity. The March 2026 core update, which completed on April 8, produced some of the sharpest ranking redistribution on record, with SE Ranking data showing 79.5% of top-three URLs changing positions during the cycle. That followed the December 2025 core update, which ran for 18 days and produced severe visibility losses for news publishers. The pattern of intensifying disruption across successive updates makes a consolidated historical reference increasingly useful for diagnosing what has changed and when.

The shift toward AI-powered search is not a future event from the perspective of Google's internal architecture. It is a decade-long process now visible to end users. DemandSphere's tracker attempts to show both the ranking enforcement history and the AI development history as a single continuous timeline rather than two separate stories. Understanding whether a ranking change coincided with a core update, a spam action, or an AI infrastructure milestone has become a necessary step in any post-mortem analysis - and a tool that has tracked that history for 25 years and surfaced it through a public API is a practical addition to that process.

The tracker is the second tool in DemandSphere Radar, the company's search intelligence initiative. The first, the AI Frontier Model Tracker, was released the previous week. Additional features for the Radar toolset are described as in development.

Timeline

  • 2000-2002: First documented Google algorithm changes, tracked in DemandSphere's historical database, cover the foundational era of search ranking.
  • November 2003: Florida update, the first widely named algorithm change, alters ranking signals for commercial queries.
  • April 2012: Penguin update targets link spam and manipulative link-building practices.
  • 2013: Hummingbird rewrites the core algorithm for semantic understanding; DemandSphere identifies this as Google's first major AI integration into search ranking.
  • October 2015: RankBrain adds machine learning to handle novel queries, later confirmed as Google's third most important ranking signal.
  • June 2017: Google Research publishes "Attention Is All You Need," establishing the Transformer architecture used in virtually all modern large language models.
  • October 2019: BERT is applied to 100% of English queries; Google calls it the biggest improvement in five years.
  • 2021: MUM, described as 1,000x more powerful than BERT, adds multimodal and multilingual AI to ranking.
  • March 5 - April 19, 2024: March 2024 core update runs 45 days - the longest rollout in the tracker's dataset - absorbing the helpful content system into core ranking.
  • May 14, 2024: AI Overviews launch, placing generative AI output directly into Google Search results.
  • March 25, 2025: AI Mode classified as complete in DemandSphere's tracker, marking fully conversational search with cited sources.
  • March 13 - 27, 2025: March 2025 core update completes in 14 days.
  • June 30 - July 17, 2025: June 2025 core update runs 16 days with significant ranking volatility.
  • August 26 - September 21, 2025: August 2025 spam update runs 26 days.
  • December 11 - 29, 2025: December 2025 core update runs 18 days, producing severe publisher visibility losses.
  • March 24, 2026: March 2026 spam update goes live at 12:18 PDT; completes in approximately 19.5 hours - the fastest spam update on record.
  • March 27 - April 8, 2026: March 2026 core update runs 12 days; SE Ranking data shows 79.5% of top-three URLs shifted positions.
  • April 18, 2026: DemandSphere publishes the Google Algorithm & AI Search Update Tracker, covering 173 entries from 2000 to 2026 under CC BY-NC 4.0, with a public JSON API.

Summary

Who: DemandSphere, an AI search visibility platform operated by Ginzamarkets, Inc., founded and led by Ray Grieselhuber.

What: The release of the Google Algorithm & AI Search Update Tracker, a free public tool documenting 173 confirmed Google Search ranking updates and AI research milestones from 2000 to 2026, available with a public JSON API under a Creative Commons license.

When: The tracker was published on April 18, 2026, the same week DemandSphere bundled it with the AI Frontier Model Tracker - released the previous week - into the DemandSphere Radar toolset.

Where: The tracker is hosted at demandsphere.com/research/demandsphere-radar/algorithm-update-tracker. The JSON API is at /api.json. Data sources include Google's Search Status Dashboard (status.search.google.com), DemandSphere's internal annotation database covering 20-plus years, and Google Research publications.

Why: DemandSphere built the tool to answer a recurring customer question - whether observed ranking changes reflect an algorithm enforcement action or something else - and released it publicly after determining that no existing resource combined ranking update history with AI milestone history on a single timeline with a live data feed and an API.

Share this article
The link has been copied!