YMYL, short for Your Money or Your Life, is the label Google gives to subjects on which a bad web page can do real damage: medical symptoms, investment decisions, emergency procedures, voting rules. Pages on those subjects are held to stricter standards of accuracy and trust by the human raters who grade Google's search results and, by Google's own account, by its ranking systems. The category exists because a search engine that treats a query about heart attack symptoms the way it treats one about music award winners will eventually send someone to harmful advice. It is not a penalty or a setting visible in any Google tool, but a classification of topics that changes how every other quality judgement is weighted.

How a topic qualifies

The working definition sits in section 2.3 of Google's Search Quality Rater Guidelines, a 182-page manual whose current edition is dated September 11, 2025. A topic qualifies when content about it could significantly affect the health, financial stability or safety of people, or the welfare of society.

There are two routes in. Some topics are dangerous in themselves, such as self-harm, criminal acts or violent extremism. Others become dangerous only when the information is wrong; the manual cites heart attack symptoms, investing, earthquake safety and who can vote. Raters then sort the possible damage into four categories:

  • Health or Safety: mental, physical and emotional health, and physical or online safety.
  • Financial Security: anything that could damage a person's ability to support themselves and their family.
  • Government, Civics & Society: groups of people, issues of public interest, trust in public institutions, and election and voting information.
  • Other: a residual category for topics that could hurt people or society.

Classification is a spectrum, not a switch. A table in the guidelines contrasts tsunami evacuation routes (clear YMYL) with a weather forecast (possibly YMYL) and music award winners (not YMYL). Advice on when to go to an emergency room is clear YMYL; advice on replacing a toothbrush is borderline. Two tests settle unclear cases: would a careful person seek out experts to avoid harm, or would most people be content to ask friends? Many or most topics, the manual says, are not YMYL.

What changes once a topic is classified

In the rating workflow the label adds no separate score; it raises the bar on existing ones. Raters first check whether a page's purpose is harmful or deceptive, then weigh main content, reputation and E-E-A-T (experience, expertise, authoritativeness and trust). On YMYL topics, accuracy means consistency with "well-established expert consensus" and reputation is judged by what experts in the field say.

The floor moves too. A YMYL page that handles sensitive data while saying nothing about who runs the site is rated Lowest, as is a YMYL page whose creator plainly lacks expertise. Claims that lemons cure cancer, or that lottery tickets are a sound retirement plan, appear as examples of harmfully misleading information. Expertise is not always demanded. Section 3.4.1 accepts first-hand experience for some content, such as a forum thread on coping with liver cancer treatment, while reserving treatment options and life-expectancy figures for experts.

In live search, Google's own descriptions imply a three-step sequence: a query arrives, systems detect whether it touches a YMYL topic, and signals associated with authority and trust then carry more weight in ranking. Google's Search Central documentation states that its systems give "even more weight" to content aligned with strong E-E-A-T on such topics. A February 2019 Google white paper on disinformation said the same of queries its algorithms detect as YMYL, according to Search Engine Land. Internal documentation leaked in 2024 contained two integer fields, ymylHealthScore and ymylNewsScore, each described as storing the output of a YMYL classifier.

Individual rater scores do not move pages; Google treats them as feedback. Google alone defines the category, and the roughly 16,000 contract raters described in a November 2023 Google overview apply it to samples. Publishers, health and finance brands and their agencies sit on the receiving end.

From page type to potential harm

YMYL was already present in version 4.1 of the rater guidelines, dated March 29, 2013, according to Jennifer Slegg, writing in Search Engine Watch that November. The first definition covered pages that could affect a person's "current or future well being". The examples covered pages collecting identification numbers, any page used for a monetary transaction, medical, legal and financial information, and advice on major life decisions such as parenting or buying a home. Read literally, most online shops qualified.

Expertise, authoritativeness and trustworthiness (E-A-T) were tied to the category around 2014, with accounts differing on the revision. Wider attention followed Google's core update of August 1, 2018, which Barry Schwartz of Search Engine Roundtable nicknamed "Medic" after many health sites reported losses. Google rejected the framing. Pedro Dias, a former Google employee, relayed Gary Illyes's view that the update was not built to target YMYL sites and that its effect on them was "a happy coincidence". The analyst Glenn Gabe catalogued more than 210 affected sites, many outside YMYL. Marie Haynes, a search engine optimisation (SEO) consultant, dates the shift earlier, to an undocumented change in February 2017 that hit YMYL content first.

The definition was rewritten on July 28, 2022. That edition replaced lists of page types with harm-based categories and the spectrum table. Experience joined the framework in December 2022. The September 2025 edition renamed the society category Government, Civics & Society and named election and voting information explicitly.

Why it matters for marketers

YMYL sectors move hardest when Google updates its ranking systems. Gabe saw "a ton of volatility" among health, finance and legal sites in the June 2025 core update. After the December 2025 update, established health publishers lost visibility they had held for years, something Lily Ray, an SEO consultant, called "really wild" to watch. A separate study of 100,000 keywords found healthcare results still anchored to long-established domains.

Generative answers raise the stakes. Google handles about one billion health questions a day, a figure given at its Check Up event on March 17, 2026. AI Overviews, the generative artificial intelligence (AI) summaries atop Google's results, hold attention on these topics: in a study of 70 users, readers scrolled through 52% of the summary on health YMYL queries and 46% on financial ones, against a median of about 30% overall. Blood Cancer UK lost 53% of views on its leukaemia page over twelve months, and Index Exchange data showed health and nutrition ad opportunities down 40% to 50% in 2025, against 7% for news. For programmatic buyers, that is a shrinking pool of health inventory on the open web.

Advertising follows a parallel logic. Google kept healthcare, finance, gambling and other sensitive verticals out of ads inside AI Overviews. Healthcare's exclusion was first breached by a small US test in AI Mode, confirmed on May 20, 2026 by Ginny Marvin, Google's Ads Product Liaison.

Limitations and disputes

The first dispute is over what YMYL is. Google frames it mainly as rater guidance, yet the evidence above points to automated detection. Mark Williams-Cook's December 2024 analysis of an exposed Google endpoint described separate handling and stricter standards for YMYL queries. Site owners can neither see the classification nor appeal it.

Incumbency is the second. Standards built on reputation and expert endorsement favour established brands over smaller or patient-led sites, though the December 2025 losses at long-standing health publishers show that incumbency guarantees nothing.

Expert consensus is the third. The guidelines exempt claims that are reasonably debatable but leave raters to decide where debate ends, a task that has become politically charged since election information entered the category. Raters are told not to let their political views shape ratings.

The last concerns Google's own answers. Two weeks after AI Overviews launched, Liz Reid, then Google's vice-president of search, wrote on May 30, 2024 that for health "we launched additional triggering refinements to enhance our quality protections." In January 2026, after a Guardian investigation, Google removed AI Overviews for queries on normal ranges for liver blood tests; a spokesperson said in-house clinicians had found much of the flagged information was not inaccurate, according to TechCrunch. Vanessa Hebditch of the British Liver Trust said the removals were not "tackling the bigger issue of AI Overviews for health."

Not the same as

E-E-A-T is the quality framework itself; YMYL sets how strictly it is applied.

Sensitive verticals in Google Ads, such as healthcare and financial services, are governed by advertising policies with certification requirements. They overlap with YMYL subjects but apply to ads, not organic pages.

Special Ad Categories on Meta restrict audience targeting for housing, employment, financial products and services, and social issues, elections or politics, to prevent discrimination.

Your Money or Your Life is also a 1992 personal finance book by Vicki Robin and Joe Dominguez, named after the highwayman's demand and unrelated to search.

Recent developments

Google has started addressing YMYL standards to site owners rather than raters alone. Additions to its helpful content guidance, made on October 2, 2026, state that content on YMYL topics must be highly accurate and consistent with established expert consensus. The same edit describes fabricated author profiles, including AI-generated headshots and false credentials, as deception. Ray argued on October 4 that the changes foreshadow a large ranking update. Google has drawn no such link, and as of October 2026 the rater guidelines remain the September 2025 edition.

Timeline

  • March 29, 2013: Version 4.1 of Google's rater guidelines includes a section on YMYL pages
  • November 19, 2013: Search Engine Watch publishes a detailed account of the YMYL section
  • 2014: E-A-T becomes tied to YMYL pages, with accounts differing on the revision
  • November 19, 2015: Google publishes the full rater guidelines
  • February 2017: An undocumented algorithm change affects YMYL content first, according to Marie Haynes
  • August 1, 2018: A broad core update, later nicknamed Medic, hits many health and medical sites
  • August 19, 2018: Gary Illyes says the update was not designed to target YMYL sites, according to Pedro Dias
  • February 2019: Google's disinformation white paper describes heavier weighting of E-A-T on YMYL queries
  • July 28, 2022: The rater guidelines redefine YMYL around potential harm and add the spectrum table
  • December 15, 2022: Experience is added, creating E-E-A-T
  • May 2024: AI Overviews launch in the US; leaked documentation exposes ymylHealthScore and ymylNewsScore fields
  • May 29, 2024: Google confirms the leaked documents are authentic
  • May 30, 2024: Liz Reid announces additional triggering refinements for health in AI Overviews
  • December 15, 2024: Mark Williams-Cook discloses endpoint data showing separate YMYL query handling
  • June 2025: The June core update produces heavy volatility among health, finance and legal sites
  • September 10, 2025: A Frankfurt court sets strict accuracy expectations for health content in AI Overviews
  • September 11, 2025: The rater guidelines rename the society category Government, Civics & Society and add election and voting information
  • December 2025: The December core update cuts the visibility of established health publishers
  • January 11, 2026: Google removes AI Overviews for liver test queries after a Guardian investigation
  • March 17, 2026: Google says it handles about one billion health questions a day
  • May 20, 2026: Google confirms a small US test of healthcare ads in AI Mode
  • August 2026: Blood Cancer UK reports a 53% fall in views of its leukaemia page
  • October 2, 2026: Google's helpful content guidance gains YMYL accuracy language aimed at site owners

Summary

Who: Google defines YMYL and applies it through its ranking systems and roughly 16,000 contract quality raters. Health, finance, legal and civic publishers, the brands behind them and their SEO agencies are most exposed, along with advertisers buying inventory on those sites.

What: A classification of topics where inaccurate or harmful content could significantly affect health, financial stability, safety or the welfare of society, sorted into Health or Safety, Financial Security, Government, Civics & Society, and Other. Pages on clear YMYL topics face stricter standards of accuracy, reputation and trust.

When: Present in Google's rater guidelines by March 29, 2013, tied to E-A-T around 2014, redefined around harm on July 28, 2022, expanded to election and voting information on September 11, 2025, and written into site-owner guidance in October 2026.

Where: In Google Search worldwide, through rater evaluations and, by Google's account, ranking systems that weight authority and trust more heavily on YMYL queries, with knock-on effects in AI Overviews, AI Mode and Google's ad placement rules.

Why: Search results on medical, financial and civic questions can cause real harm when wrong. Google uses YMYL to demand more of those pages, though the classification is invisible to site owners, favours established sources and sits awkwardly beside Google's own AI-generated answers on the same topics.