IAB Content Taxonomy vs Ad Product Taxonomy: what each one labels and when to use it
Both lists come from IAB Tech Lab and both are full of category names, so they are easy to confuse. The difference is simple once you see it: one describes the place where an ad runs, the other describes the thing being advertised.
Content Taxonomy
- Labels pages, sites, apps, videos, podcasts
- Used for contextual targeting and suitability
- Current release: version 3.1
Ad Product Taxonomy
- Labels creatives and advertisers' offers
- Used for ad blocking rules and competitive separation
- Current release: version 2.0
Two questions, two vocabularies
An ad impression has two sides. The publisher side knows what the content is about. The advertiser side knows what product the creative promotes. Each side needs its own words, because the same subject can be fine on one side and a problem on the other.
Same subject, different meaning
A recipe site about cocktails is Content: Food & Drink. An ad for a beer brand is an alcohol product. A publisher may welcome the first and restrict the second.
Products are more specific
Content subjects are written for readers' interests. Product categories are written for what is bought, with finer splits where rules and regulation differ.
Different people own them
Editorial and ad operations teams care about content labels. Brand, legal and sales teams care about product labels and what may be shown next to what.
Side by side comparison
| Aspect | Content Taxonomy | Ad Product Taxonomy |
|---|---|---|
| Describes | The subject of a page, site, app, video, podcast or CTV programme | The product or service promoted in an ad creative |
| Typical label | Travel > Travel Locations, Sports > Golf | A product type such as a consumer loan, an airline ticket, a beer |
| Applied by | Publishers, SSPs, contextual and verification vendors, data providers | Advertisers, agencies, DSPs, creative review teams |
| Main jobs | Contextual targeting, brand suitability, inventory packaging, research | Ad blocking rules, ad quality review, competitive separation, reporting on demand |
| Versions | 1.0, 2.0, 2.1, 2.2, 3.0 and 3.1 (December 2024) | 1.0 and 1.1 (2022), then 2.0, the current release |
| Shape | Four tier columns with IDs, parents and a flag column | A tiered list with IDs and parents; commonly blocked categories sit at the top level in 2.0 |
| Sister lists | Both sit alongside the IAB Audience Taxonomy, which describes groups of people rather than content or products. | |
For a tour of the content side, see our guide to IAB Content Taxonomy 3.1 structure and categories.
Where each taxonomy appears in a programmatic auction
In a typical OpenRTB auction both lists travel in the same transaction, but in different places and in different directions.
The site, section or page carries Content Taxonomy codes describing what it is about.
The request lists product categories the publisher will not accept, for example certain financial or adult products.
The DSP matches the content codes against the campaign's contextual include and exclude lists.
The bid states the product categories of the creative so the seller can check its rules.
The exchange compares declared product categories with the block list and the content with any separation rules.
| OpenRTB field | Which taxonomy it usually carries | What it says |
|---|---|---|
| site.cat, app.cat | Content | The subjects of the site or app as a whole |
| sectioncat, pagecat | Content | The subjects of the current section or page |
| bcat (in the request) | Ad Product | Product categories the seller blocks for this impression |
| cat (in the bid) | Ad Product | Product categories of the creative being offered |
| cattax | Either | Which taxonomy and version the neighbouring category codes belong to |
Older integrations often used Content Taxonomy 1.0 codes for everything, including ad categories. That habit is the reason IAB Tech Lab publishes mappings between the content and product lists, and why declaring the taxonomy with each list matters. Our programmatic advertising and RTB use case covers the data side in more depth.
When to use which
Use the Content Taxonomy when you...
- Choose where a campaign should appear by subject
- Build domain or page include lists for contextual targeting
- Exclude sensitive subjects for brand suitability
- Package a publisher's inventory into sellable segments
- Describe a website for research, enrichment or market sizing
- Replace cookie-based segments with context signals
Use the Ad Product Taxonomy when you...
- Tell sellers what a creative promotes
- Block types of ads on your own inventory
- Keep competing brands apart in the same ad break
- Apply regulated product rules, such as alcohol or gambling
- Report demand by product type across advertisers
- Review creatives before they go live
| Scenario | Taxonomy | Why |
|---|---|---|
| A bank wants to appear on personal finance sites | Content | The decision is about where the ad runs |
| A children's app refuses ads for gambling | Ad Product | The rule is about what is advertised |
| A brand avoids news about disasters | Content | Suitability depends on the content subject |
| A streaming service separates two car brands in one break | Ad Product | Separation compares the products in the ads |
| A beer brand wants sports sites but not sites for minors | Both | Content picks sports, product rules handle the alcohol restriction |
| An analyst sizes the market of travel websites | Content | The question is about what the sites are |
| A DSP reports spend by advertised industry | Ad Product | Spend follows the products being sold |
How the two lists relate to each other
The lists overlap in vocabulary, but they are not two views of the same tree. A mapping between them is a helpful bridge, not a perfect translation.
Mappings exist
IAB Tech Lab has published mappings between Content Taxonomy 1.0 and Ad Product Taxonomy 2.0 in both directions, and later work on a Content 2.1 to Ad Product 2.0 mapping.
One to many is normal
One content subject, such as Personal Finance, relates to many product types: cards, loans, insurance, investing apps. The reverse is also true.
Some rows have no partner
Plenty of content subjects, such as Genres or Pop Culture, have no product equivalent, and some regulated products have no single content home.
Keep them in separate fields
Store content and product labels in separate columns with their own version numbers. Mixing them in one field causes silent errors later.
Use mappings for defaults
A mapping is a good way to propose a starting block list or a starting contextual list. A person should still review the result for the campaign at hand.
Audience is the third axis
The Audience Taxonomy describes people. Context data can stand in for some audience groups without tracking, which is where personas inferred from content help.
Why there are two lists at all
For years the industry used one list for both jobs. Splitting them was a response to the limits of that approach.
| Stage | What happened | Why it mattered |
|---|---|---|
| One shared list | The original Content Taxonomy 1.0 codes were used to describe pages and also to describe ads and blocked ad types. | Simple, but a subject list is a poor fit for products, and regulated products had no precise codes. |
| Deeper content trees | Content Taxonomy 2.x and then 3.x grew into detailed hierarchies built for describing content. | The content list became better at its own job and even less suited to describing ads. |
| A dedicated product list | IAB Tech Lab launched the Ad Product Taxonomy in 2022, with versions 1.0 and 1.1 that year. | Ads could finally be described by what they sell, including categories such as foods high in fat, sugar or salt. |
| Product list 2.0 | Version 2.0 added rows and moved commonly blocked categories to the top level. | Block lists became easier to build and to audit. |
| Bridges between them | Mappings between Content 1.0 and Ad Product 2.0, and later between Content 2.1 and Ad Product 2.0, were published. | Teams with old rules got a practical path to the newer product list. |
Who uses which, by role
Most companies in the ad chain touch both lists, but each role leans on one of them for its daily work.
| Role | Leans on | Typical task |
|---|---|---|
| Publisher | Content Ad Product | Labels its own pages and sets which ad types it refuses |
| SSP or exchange | Content Ad Product | Passes content labels to buyers and enforces sellers' product blocks |
| DSP | Content | Matches inventory context to campaign include and exclude lists |
| Agency trader | Content | Builds contextual lists and suitability settings per campaign |
| Brand and legal team | Ad Product | Defines what its own products count as and where rules apply |
| Verification vendor | Content | Checks that ads appeared in suitable environments |
| Data provider | Content | Supplies category labels for domains, pages or apps |
| Researcher or analyst | Content | Sizes markets of websites, studies media landscapes |
Common mix-ups and how to avoid them
Blocking content when you meant products
A publisher that wants no gambling ads sometimes blocks the gambling content subject instead. That stops nothing, because the rule needs to look at the ad, not the page.
Targeting products when you meant content
A campaign set to "Travel" product codes will not find travel websites. Contextual reach comes from content codes on the inventory side.
Assuming codes are interchangeable
Two lists can contain a row with the same or a similar name. The IDs and the meaning still differ, so translate through a mapping and a review.
Leaving out the taxonomy declaration
Sending codes without stating which list and version they come from forces the receiver to guess. In OpenRTB, the cattax field exists to remove that guess.
Overblocking regulated products
Blocking a whole top-level product branch can remove harmless ads. Pick the precise rows that your policy or local rules actually require.
Ignoring sensitive content flags
Some content rows touch sensitive personal data. Using them as context is different from profiling people, but privacy teams should still sign off.
Where domain category data fits
Domain category data sits on the content side. It tells you what each website is about, before any auction happens, so you can build lists and rules in advance.
| Need | What domain data gives you | Related page |
|---|---|---|
| Contextual include lists | Every domain whose ranked IAB v3 categories match your chosen IDs, filtered by popularity | Contextual targeting for adtech |
| Suitability exclusions | Domains labelled with sensitive subjects and web filtering categories such as Adult or Gambling | Brand safety for advertising |
| Supply quality screening | Category and popularity signals that help spot thin or suspicious sites | Made for advertising sites |
| Audience without cookies | More than 1,900 personas inferred from site content | Audience segmentation |
| Publisher packaging | Consistent IAB labels for a publisher's own domains and partner sites | Publisher monetization |
The database does not label ad creatives, so it does not assign Ad Product codes. It gives the content side of the equation for 120 million domains, in IAB v3 and IAB v2, which you can pair with your own product rules. You can look up any site with the domain category checker.
Implementation tips for data teams
Version every column
Name columns with the taxonomy and version, for example content_v3 and adproduct_v2. Future you will be grateful when the next release arrives.
Keep a parent table
For each taxonomy, keep a small table from every ID to its tier 1 parent. It makes roll-up reporting quick on both sides.
Write rules in plain words first
"No alcohol ads on teen content" is clear to everyone. Translate it into product and content codes afterwards, and keep both versions in the rule record.
Test with real examples
Before switching rules on, run a set of known domains and creatives through them. Check the blocked and allowed results by hand.
Refresh content labels
Sites change subject and new domains appear every day. Content labels need regular refreshes; product labels change only when the creative does.
Keep the lists handy
Our taxonomy downloads include the IAB content lists used in our data in JSON and CSV, ready for your lookup tables.
Before you send content labels to a partner
- State the taxonomy name and version with every list
- Send IDs, with names only as a readable extra
- Say whether labels are site, section or page level
- Say how several labels per domain are ranked
- Agree how often the labels will be refreshed
Before you send product rules to a partner
- Write each rule in plain words as well as codes
- List the exact product rows, not whole branches
- Note which rules come from law and which from policy
- Give an owner and a review date for each rule
- Test the rules with real creatives before launch
Frequently asked questions
What is the main difference between the two taxonomies?
The Content Taxonomy describes what content is about. The Ad Product Taxonomy describes what an ad is selling. One is about the environment, the other is about the message.
Which one should a publisher use to block ads?
Use the Ad Product Taxonomy for rules about what may be advertised. Use the Content Taxonomy to describe your inventory so buyers can find it.
Can I still use Content Taxonomy codes for ad categories?
Many older systems did, and some still accept it. The product list is more precise for ads, and mappings exist to help move rules across.
Is the Ad Product Taxonomy tiered like the Content Taxonomy?
Yes, it uses IDs, parents and tiers. In version 2.0 the categories that are most often blocked were placed at the top level so they are easy to find.
Do I need both for contextual advertising?
For targeting, the content side is what matters. Product categories come in when sellers apply block lists or separation rules to your creatives.
Which version of each list should a new project use?
Use the current releases: Content Taxonomy 3.1 and Ad Product Taxonomy 2.0. Keep older codes only where a partner still requires them, and say which version each list uses.
Does the Audience Taxonomy replace either of them?
No. It describes groups of people, such as age bands or interests. Content and product lists stay in use alongside it, and many teams now infer audience groups from content instead of tracking.
How should I store both in a data warehouse?
Give each its own column or table, with the taxonomy name and version in the column name. Keep a parent table for each so reports can roll up to tier 1 without guesswork.
Do you sell Ad Product labels?
No. Our database labels websites, which is content data. It carries IAB v3 and v2 content categories, web filtering categories, personas and popularity for each domain.
The content side, ready for 120 million domains
Ranked IAB content categories for every domain in one licensed file, with a lookup API included. Tell us how you plan to use it and we will suggest a plan.
See plans and pricing IAB categories listIAB and IAB Tech Lab are trademarks of the Interactive Advertising Bureau. OpenRTB is a specification of IAB Tech Lab. These names are used here only to identify the standards discussed. URL Categorization Database is not affiliated with, sponsored or endorsed by IAB or IAB Tech Lab. All explanations are our own.