Tenten AIGEO
Back to Blog
Technical AEOConsideration

Article or BlogPosting Schema? Which structured data should be chosen when AI cites articles?

BlogPosting is a subtype of Article. In most cases, whichever one you choose will not affect AI citations. This article uses type inheritance, three-step decision-making method and four common mistakes to explain clearly the fields that should be really concerned about when selecting Article schema, and what situations will really make the wrong choice.

Tenten GEO TeamPublished 2026-07-125 min read
Cover of bifurcated glowing path symbolizing choice between two structured data types

For more than 90% of B2B content, choosing Article or BlogPosting has almost no impact on whether the AI engine will cite you. What determines fate is not the name of the type, but which fields are filled in the type and whether they are filled in cleanly enough. But "almost" leaves a gap: there are certain kinds of content that once the wrong type is marked, you will miss the search function that you originally had. This article separates what should be confused and what should not be confused at once.

Let me first talk about a fact that many people don’t know: BlogPosting is originally a subtype of Article. The inheritance chain of schema.org is Article → SocialMediaPosting → BlogPosting, which means that when you mark BlogPosting, you also declare "This is an Article, and more specifically, a blog post." The two share exactly the same core fields; Google's documentation directly puts the three types of Article, NewsArticle, and BlogPosting into the same specification, and the required and recommended fields are exactly the same.

Why in most cases "which one to choose" is less important than you think

When the AI engine and Google judge an article, they read the "attributes" in the structured data, not the "tags" of the type. It needs to know who wrote it, when it was published, when it was updated, which website it belongs to, and which main picture it is. These fields have the same names in Article and BlogPosting. In other words, if you change @type from Article to BlogPosting, the string of author, datePublished, dateModified, and publisher will not be changed at all, and the information read by the machine will be exactly the same.

  • headline: article title, it is recommended to be within 110 characters and match the page title.
  • author: Be sure to use a Person or Organization object, and fill in sameAs to point to a verifiable author identity.
  • datePublished and dateModified: The publication and last update time are the main basis for AI to judge the freshness of content.
  • Publisher: The publishing organization, including logo, is used to tie the article back to the brand entity.
  • image and mainEntityOfPage: the main image and the official URL of this article.

So when do you really make the wrong choice?

Where the difference bites, it's the two more specific types. The first is NewsArticle. Google has special treatment for it, and it’s one of the tickets into Top Stories and Google News. But this is a double-edged sword: if your content is not a timely news report but is labeled NewsArticle, not only will you not get news placement, but it may also be judged that the type does not match the content. On the other hand, marking a real industry news as BlogPosting is equivalent to closing the door to the news section.

The second type is TechArticle, which is used in technical documents, developer teaching, and implementation guides. It has additional fields such as proficiencyLevel and dependencies, which can clearly tell the engine what level of people this article is for and what prerequisites are required. For developer-oriented SaaS, an API teaching standard TechArticle will be more accurate than BlogPosting; and if the article has clear steps, it can also be superimposed with HowTo. The principle here is simple: choose the "accurate and most specific" type, and don't fall back on the most general one just for insurance.

A three-step decision-making method, completed in 30 seconds

  1. Is this content with a clear time limit and reporting nature? Yes → NewsArticle (provided you are truly journalistic qualified).
  2. Is this a technical document, development tutorial, or step-by-step implementation? Yes → TechArticle, with steps plus HowTo.
  3. No, then is it an article on your blog with a published date? Is → BlogPosting.
  4. If it’s an evergreen resource page, guide, or knowledge base article that doesn’t hang on the timeline → use Article.
Schematic diagram of the decision-making process for selecting Article, BlogPosting, NewsArticle, and TechArticle types
Ask about timeliness and technicality first, neither of them, then fall to BlogPosting or Article.

What the AI engine is really reading are these fields.

Choosing the right type is just a passing line. What really makes an article cited repeatedly in ChatGPT, Perplexity, and Google AI Overviews is that the fields are filled in completely and verifiably. If the author is just a string "editorial office", the engine cannot judge the authority; by adding the Person object and adding sameAs to point to LinkedIn or the author page, the E-E-A-T signal can be established. If dateModified stops at two years ago, AI will tend to use newer sources, and no matter how well you write, you will still be ranked behind.

Four mistakes more fatal than choosing the wrong type

  • Using WebPage or simply not typing is equivalent to giving up the entire Article field space.
  • author uses pure text, without Person objects and sameAs, and the authoritative signal is directly reset to zero.
  • The headline does not match the page title, or it exceeds 110 characters and is truncated.
  • Multiple contradictory types are forced into the same node, or the same set of datePublished is copied and pasted in each article.

When we do GEO audits for B2B SaaS clients, incorrect typing is rarely the main reason. A more common picture is: all articles on the site are marked with BlogPosting, but the author is plain text, dateModified has never been updated, and the publisher lacks a logo. The type is correct, but the field is empty, and the AI ​​still cannot extract information that can build trust. The response rate for repairing this level is usually much higher than struggling over type names.

The type determines how the engine classifies you, and the field determines whether the engine trusts you. If you choose the wrong category, you will miss one slot at most. If the column is empty, the entire article will lose weight in the eyes of AI.Tenten GEO

So, I would suggest this

Use BlogPosting for general blog articles, Article for evergreen resources, NewsArticle for news, and TechArticle for technical teaching. Choose the most accurate and specific one, and then put all your efforts into the completeness of the fields. Don’t spend more than thirty seconds on type names. If you want to know whether the structured data on your website is of the wrong type or if the fields are empty, you can book a 30-minute GEO diagnosis. We will use your actual articles to point out the fields that should be filled first.

Frequently asked questions

What is the difference between Article and BlogPosting schema?
BlogPosting is a subtype of Article. They share the same core fields and are processed by Google using the same specification. The difference is only in the specificity of the semantics: BlogPosting clearly indicates that this is a blog article, while Article generally refers to a general article.
Will the AI engine make it easier to cite because I choose BlogPosting?
No. The AI ​​engine reads the author, datePublished, dateModified, and publisher fields instead of the type name. Filling out the fields completely and verifiably is far more important than which type you choose.
When should you use NewsArticle or TechArticle?
Use NewsArticle for timely news reports to compete for Top Stories placement; use TechArticle for technical documents and development tutorials, and add HowTo if there are clear steps. The principle is to choose the accurate and most specific type.

READY WHEN YOU ARE

How visible is your brand in AI answers?

In a 30-minute GEO diagnostic, we use real prompts to identify your visibility gaps across major AI engines and show you what to fix first.

Book a 30-minute diagnostic