What was the GARM brand safety framework, and how do you cite it now?
Part of Brand suitability
The short answer
The GARM brand safety framework was the Brand Safety Floor and Suitability Framework published by the Global Alliance for Responsible Media, a World Federation of Advertisers initiative. It graded sensitive content categories into a floor and three risk levels. The WFA discontinued GARM in August 2024, so the framework is now a legacy reference, not an operating standard or certification.
Key takeaways
- GARM was a voluntary, cross industry initiative the World Federation of Advertisers created in 2019 and discontinued in August 2024.
- The framework grew out of the 4A's Advertiser Protection Bureau floor of September 2018 and was released with that bureau in September 2020.
- Its category list survives in documents that are still maintained: IAB Tech Lab's Sensitive Topics branch and TAG's Brand Safety Certified guidelines.
- Phrases such as GARM compliant or GARM certified describe something that no longer exists and should be rewritten.
What was GARM?
Definition
- GARM (Global Alliance for Responsible Media)
- A voluntary, cross industry initiative of the World Federation of Advertisers on digital safety, created in 2019 and discontinued in August 2024.
The WFA's own closing announcement (opens in a new tab) gives the origin story. GARM was set up in 2019 after the Christchurch shootings, which the attacker livestreamed on a social platform, and after brands found their ads beside illegal or harmful content. It was not a regulator, auditor or certification body. Its output was shared definitions and work on transparency in social media ad placements, offered so that each advertiser could make its own decisions about where its money went.
GARM did not start from nothing. The Media Rating Council's September 2018 supplement (opens in a new tab) on content level brand safety already pointed to the 4A's Advertising Assurance Brand Safety Floor Framework, issued that month, which it said held baseline floor definitions within 13 categories rooted in the industry's dirty dozen list. The IAB's December 2020 guide (opens in a new tab) describes GARM as the collective that later partnered with the 4A's bureau to release an updated floor and suitability framework. A 2019 brief citing the 4A's floor and a 2022 brief citing GARM can be describing close relatives of the same list.
What did the GARM Brand Safety Floor and Suitability Framework contain?
The framework is a grid. Each row is a sensitive content category. The first column describes the floor, which the copy hosted by the Brand Safety Institute (opens in a new tab) labels content “not appropriate for any advertising support”. The next three columns describe high, medium and low risk content that is “appropriate for advertising supported by enhanced advertiser controls”. Its stated goals were consistent categorization, transparency and clarity on exceptions for responsible speech.
The September 2020 edition had 11 categories. IAB Tech Lab's implementation guide for Content Taxonomy 2.2 (opens in a new tab) says the floor and suitability concepts in that taxonomy version are based on the framework released in September 2020 by GARM with the 4A's Advertiser Protection Bureau, and lists the 11 categories it imported. The copy online today has 12, with misinformation as the last row. The wording also drifted slightly between documents.
| Framework copy (Brand Safety Institute) | Content Taxonomy 3.1, Sensitive Topics | Taxonomy 3.1 ID |
|---|---|---|
| Adult & Explicit Sexual Content | Adult & Explicit Sexual Content | Rm3SiT |
| Arms & Ammunition | Arms & Ammunition | avbNf2 |
| Crime & Harmful acts to individuals and Society, Human Right Violations | Crime & Harmful Acts to Individuals, Society & Human Right Violations | XtODT3 |
| Death, Injury or Military Conflict | Death, Injury, or Military Conflict | I4GWl6 |
| Online piracy | Online Piracy | mm3UXx |
| Hate speech & acts of aggression | Hate Speech and Acts of Aggression | HxqYV1 |
| Obscenity and Profanity, including language, gestures, and explicitly gory, graphic or repulsive content | Obscenity and Profanity | j9PaO9 |
| Illegal Drugs, Tobacco, ecigarettes, Vaping, Alcohol | Illegal Drugs, Tobacco, eCigarettes, Vaping, Alcohol | pg0WhF |
| Spam or Harmful Content | Spam or Harmful Content | 6i4dB6 |
| Terrorism | Terrorism | 8FD8nI |
| Debated Sensitive Social Issue | Debated Sensitive Social Issues | Z7rJBM |
| Misinformation | No matching category | None |
The IDs come from the Content Taxonomy 3.1 file (opens in a new tab), where all 11 sit under the parent Sensitive Topics, ID v9i3On. The framework itself had no IDs. An ID can travel in a data feed; a paragraph of framework prose cannot.
Why did GARM end?
The World Federation of Advertisers discontinued GARM in August 2024. Its Brand Safety Floor and Suitability Framework is a legacy reference, not an operating standard or certification. In its announcement (opens in a new tab) the WFA called GARM a small, not for profit initiative and said recent allegations that misconstrued its purpose and activities had caused a distraction and significantly drained its resources and finances. That is the WFA's own account, not a finding by a court or regulator.
The wider argument about coordinated ad buying did not end with GARM. In September 2025 the FTC's press release (opens in a new tab) on its final consent order with two merging agency groups said the complaint alleged that agencies coordinated, including through industry associations, on decisions not to advertise on certain websites and apps. The order bars withholding spend from a publisher over its political or ideological viewpoint unless an advertiser customer gives express and individualized direction. For a buyer, the lesson is authorship: a shared list is a description, and each exclusion should trace back to the advertiser's own decision.
Can a brand still use the GARM framework?
Yes, as vocabulary. A brand can describe its floor and risk levels with the framework's words, as long as the policy names the copy it quotes. What it cannot do is claim that following the framework proves anything today. There is no GARM membership to hold, no review to pass and no update to follow.
Two maintained documents still carry the list and are safer citations when a contract or request for proposal needs a current reference. TAG's Brand Safety Certified guidelines (opens in a new tab), version 5.0 dated July 2026, require certified buyers, sellers and intermediaries to use a TAG recognized content taxonomy to avoid the harmful content floor; Appendix B prints the 11 Sensitive Topics from Taxonomy 2.2 and the 12 category framework. IAB Tech Lab keeps the 11 categories in Content Taxonomy 3.1, the version a data feed can name. Neither document turns the framework into a standard a campaign can comply with.
What should replace GARM references in a brand policy?
Replace the authority, not the vocabulary. A policy that says follow GARM can stand for three things: a category list, a floor the brand does not fund, and a tolerance per category above it. Each needs a source that exists today and an owner inside the brand. The brand suitability hub covers choosing tolerances; the steps below cover only the rewrite.
Five steps to retire a GARM reference without changing what the policy means
- Find every mention. Search the policy, briefs, insertion order templates and partner questionnaires for GARM, the 4A's floor and dirty dozen.
- Say what each mention does. Mark it as a category list, a floor, a tolerance, a partner requirement or a compliance claim. Split sentences that do two jobs.
- Point the category list at a maintained source. Name Content Taxonomy 3.1 Sensitive Topics with IDs, and keep the framework copy as a dated note for risk level wording.
- Write the floor and tolerances in the brand's words. Quote the framework where it helps, and record who approved each line and when.
- Delete status claims. Remove GARM compliant and similar labels. Where a partner standard is needed, name the TAG seal or the MRC accreditation and the version checked.
| Old wording | Problem | Replacement |
|---|---|---|
| All placements must be GARM compliant. | No one can check or grant compliance. | All placements must clear the brand's floor in section 2, approved by the marketing director on a stated date. |
| Partners will follow the latest GARM framework. | There will be no later version. | Partners will classify content with IAB Tech Lab Content Taxonomy 3.1 Sensitive Topics and report the version used. |
| We use the GARM medium risk setting for news. | The risk level is borrowed; the decision is the brand's. | The brand accepts medium risk news reporting in these categories, using the archived framework's wording as a description. |
| Our vendor is GARM certified. | GARM did not certify vendors. | Our partner holds the TAG Brand Safety Certified seal, checked on TAG's website on a stated date. |
What does a GARM reference rewrite look like on a real account?
Fictional example
An Alaska seafood brand retires a GARM clause
A fictional Anchorage producer of smoked salmon and canned seafood moves its media account to a new agency. The inherited 2022 policy says every placement must be GARM compliant, with GARM low risk settings in all categories. The planner runs the five steps and finds four mentions across the policy, the insertion order template and a partner questionnaire.
The category list now cites Content Taxonomy 3.1 Sensitive Topics by ID, with the archived framework copy kept as dated background. The client notices that low risk everywhere would have shut out seafood industry news about vessel accidents, which its customers follow closely. The new policy accepts factual reporting on death and injury at sea and keeps a strict line on graphic imagery. The GARM certified question becomes a question about the TAG seal. The marketing manager signs each section, and the old clause stays in an appendix marked superseded.
How does PlacementGraph use the GARM framework today?
What should a policy review check before it cites the framework?
Framework citation check
- The policy names the exact copy of the framework used, where it is hosted and the date it was retrieved.
- The policy says the framework is a legacy reference and that GARM was discontinued in August 2024.
- Category labels that must travel to partners cite Content Taxonomy 3.1 Sensitive Topics, with IDs.
- Misinformation is either adopted as a brand item with an owner or left out on purpose, since the taxonomy has no such category.
- No sentence claims GARM compliance, certification or approval.
Run this check whenever the policy changes hands; old references can survive agency moves because they look official. The brand safety policy guide shows where these clauses sit.
Frequently asked questions
Should old campaign reports that mention GARM be rewritten?
No. Keep historical reports as they were, because they describe the rules that applied at the time. If a report could be mistaken for a current assurance, add a dated note saying GARM was discontinued in August 2024 and pointing to the policy that replaced the reference.
Did the end of GARM change any Google Ads settings?
No. Google Ads content suitability uses Google's own controls, such as the Maximum, Moderate and Limited inventory types and its sensitive content exclusions. They are not GARM settings and carry Google's own definitions, so map each line of a policy to them directly and note which surfaces each control reaches.
Were the GARM framework's categories meant only for websites?
No. The framework copy says platforms could align monetization policies with it, ad tech providers could build it into targeting and reporting, and agencies and marketers could use it for investment and suitability decisions. The WFA also credited GARM with work on social media ad placements.
Why do category names differ slightly between documents?
Each document retyped the list. The framework copy says Debated Sensitive Social Issue, Taxonomy 2.2 used Sensitive Social Issues, and Taxonomy 3.1 says Debated Sensitive Social Issues. Quote the label from the source a policy actually cites, and use the taxonomy ID when the label must match a partner's data.
Sources
- WFA discontinues GARM (opens in a new tab)World Federation of Advertisers
- Brand Safety Floor + Suitability Framework (opens in a new tab)Brand Safety Institute
- Implementation Guide for Brand Suitability with IAB Tech Lab Content Taxonomy 2.2 (opens in a new tab)IAB Tech Lab
- Content Taxonomy 3.1 (TSV file) (opens in a new tab)IAB Tech Lab
- Enhanced Content Level Context and Brand Safety, Ad Verification Supplement (September 2018) (opens in a new tab)Media Rating Council
- Brand Safety and Suitability Guide (December 2020) (opens in a new tab)IAB
- Brand Safety Certified Guidelines, version 5.0 (opens in a new tab)TAG
- FTC alters final consent order in response to public comments (opens in a new tab)Federal Trade Commission
Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.