Understanding Patent Kind Codes: Navigating Through Different Stages of Patent Documentation

When you look up a patent document – whether on WIPO’s patent portal, the USPTO database, or EPO’s Espacenet – you’ll notice a short alphanumeric suffix sitting right after the patent number. Something like B1, A2, or C1. These are called kind codes, and they’re far more informative than they look. A kind code instantly tells you what type of patent document you’re dealing with and at what stage of the patenting process that document was published. For patent searchers, IP professionals, and law students, misreading a kind code can mean the difference between citing a pending application and a granted patent – two very different things legally.

What are patent kind codes? 🔗

A kind code, formally known as a WIPO Standard ST.16 code, is a standardized alphanumeric identifier printed on patent documents by intellectual property offices worldwide. It consists of a letter – and often a number – appended to the end of a patent document number. As per the standard, these codes serve two functions: identifying the kind of patent document (e.g., an application, a granted patent, a reissued patent, or a corrected publication) and the level of publication (first publication, second publication, corrected publication, etc.).

For instance, it is recommended that USPTO documents be identified by three elements: the two-character country code (US), the patent or publication number, and the WIPO ST.16 kind code – for example, “US 7,654,321 B1” for a granted U.S. patent or “US 2003/1234567 A1” for a published patent application. That small suffix at the end carries significant legal weight.

Kind codes are not cosmetic. They are the primary tool for correctly classifying patent documents in prior art searches, citation lists, examination reports, and litigation. Without them, a researcher cannot tell from the document number alone whether they’re looking at something that has been granted or merely applied for.

The international standard behind kind codes: WIPO ST.16 🔗

Using the codes recommended in WIPO Standard ST.16, patent professionals can distinguish different types of patent documents – such as patent applications, granted patents, utility models, and more – published by various IP offices in different languages. The standard is maintained by WIPO and is part of a broader suite of industrial property documentation standards that also includes ST.3 (country codes) and ST.9 (INID codes for bibliographic data).

Under ST.16, the letter groups are structured hierarchically: the letter “A” generally denotes the first level of publication, “B” the second level, and “C” the third level. This tiered structure means that as a patent moves through the system – from application to grant to any post-grant corrections – the kind code evolves with it, creating a traceable documentary history of the patent’s lifecycle.

It’s important to note that while ST.16 provides the international framework, each national or regional patent office has discretion in how it implements the codes. This is why the same letter – say “A1” – can mean slightly different things at the EPO versus WIPO versus the USPTO. The standard harmonises the logic, not always the meaning.

Kind codes at major patent offices 🔗

WIPO and PCT applications 🔗

Under the Patent Cooperation Treaty (PCT), which allows inventors to file a single international application covering multiple countries, WIPO publishes applications in its International Bureau. For PCT international applications, the kind codes include A1 for an application published with an international search report (ISR), and A2 for an application published without an ISR. There is also an A3 code, which WIPO’s PCT Gazette uses for a separate publication of the international search report when it wasn’t ready at the time the application was first published.

An important distinction: a PCT application published by WIPO does not, by itself, become a granted patent. It must enter the national phase in each designated country for examination and potential grant. This is why WIPO kind codes focus primarily on the publication status of the application rather than tracking through to a grant stage.

USPTO (United States Patent and Trademark Office) 🔗

On January 2, 2001, the USPTO began printing WIPO Standard ST.16 kind codes on each of its published patent documents. Before this date, the USPTO only issued one type of document per patent (the granted patent), so no kind code system was necessary. The introduction of pre-grant publications changed that.

The current USPTO kind codes include: A1 for a patent application pre-grant publication; A2 for a republication of a pre-grant application; A9 for a corrected pre-grant publication; B1 for a granted patent where there was no previously published pre-grant publication; B2 for a granted patent where a pre-grant publication did exist; and C1, C2, C3 for reexamination certificates.

The America Invents Act of 2011 (AIA) further expanded the USPTO’s kind code list by introducing new post-grant proceedings such as inter partes review (IPR) and post-grant review (PGR), each generating their own unique publication documents and corresponding codes.

EPO (European Patent Office) 🔗

The EPO operates a well-structured kind code system that tracks a patent document through multiple stages of the European patent process. At the EPO, “A1” is used for European patent applications published with the European search report; “A2” for applications published without the search report; “A3” for the search report published separately; and “B1” for European patent specifications, i.e., granted patents.

The A2/A3 split is particularly notable. When an examiner hasn’t completed the search report by the time an application is published, the application goes out as A2. Later, when the search report is ready, it’s published separately as an A3 document. Together, these two documents form a complete picture of the application and the examiner’s prior art findings. For anyone doing freedom-to-operate research or patentability analysis on a European patent, checking both the A2 and A3 documents is essential.

Japan Patent Office (JPO) 🔗

The Japan Patent Office uses kind codes including B1 and B2 for publications of granted patents. Japan also uses “A” codes for unexamined applications that are open to public inspection – a practice sometimes referred to as the “kokai” publication system in Japanese IP law. Understanding JPO kind codes becomes relevant for Indian professionals dealing with technology transfer agreements or prior art searches in electronics, automotive, and pharmaceutical sectors where Japanese IP is substantial.

Common kind codes: a practical reference 🔗

Across jurisdictions, the pattern generally holds: A-type codes signal applications (pre-grant), B-type codes signal granted patents, and C-type codes typically relate to post-grant certificates or corrections. Common kind codes across patent offices include: A1 (published application with a search report), A2 (published application without a search report), A3 (separate publication of the search report), A8 (corrected application publication), A9 (complete reprint of an application publication), B1 (granted patent), B2 (amended or new patent specification after opposition), B3 (granted patent after limitation procedure), and B8 (corrected title page of a granted patent document).

The challenge is that these codes are not universal in meaning. The WIPO Standard ST.16 standardizes kind codes for global consistency, making it easier to analyse international patents – but patent databases like USPTO and Espacenet remain essential for verifying the specific meaning of a kind code in each jurisdiction. Always cross-check the kind code against the issuing office’s published table before drawing legal conclusions.

Why kind codes matter for patent searching and legal analysis 🔗

Accurate prior art identification 🔗

In a patent examination or invalidation proceeding, the type of document matters as much as its content. A published application (A1) discloses an invention but carries no legal presumption of validity – the claims may still be amended or rejected. A granted patent (B1 or B2) carries enforceable claims. All nonpatent literature documents are given the code “N,” and numerical designations are sometimes found along with letter codes on published documents – examiners are instructed to use numerical codes only if they appear on the actual document. Misidentifying an A-type document as a B-type in a legal brief or search report is a substantive error with real consequences.

Citation in examination reports and litigation 🔗

The recommended elements for identifying a patent in a citation include the office that issued the document (by two-letter country code per WIPO ST.3), the document number, and the kind of document using the appropriate ST.16 symbol. In patent examination, examiners are required to cite prior art with this level of specificity. In litigation, identifying the exact document – and its publication stage – can determine whether a piece of prior art pre-dates the priority date of the patent being challenged.

Building patent families 🔗

Publication kind codes identify different publication stages of a single application – including the initial publication, search reports, granted patents, translations, and amendments or corrections – and together these constitute a “domestic family” for that application. When doing patent landscaping or freedom-to-operate analysis, tracking all documents in a domestic family through their kind codes gives a complete picture of how the patent evolved from filing to grant, including any claims that were narrowed or dropped along the way.

Kind codes and the Indian patent landscape 🔗

India’s patent system is governed by the Patents Act, 1970 and administered by the Indian Patent Office (IPO), which publishes patent documents in the Official Journal of the Patent Office. India is a member of the Patent Cooperation Treaty, so PCT applications designating India are published by WIPO with standard kind codes before entering the national phase at the IPO for examination.

For Indian law students and practitioners, this creates a dual system to navigate: WIPO kind codes on PCT applications and the IPO’s own publication practices for domestic filings. When researching Indian patents, always verify whether the document you’re looking at is a PCT application published by WIPO (kind code from WIPO’s system) or a directly filed Indian application published by the IPO. The filing route affects which kind code table applies and, more critically, what legal rights have or haven’t yet been conferred.

Indian patent searches are frequently conducted on platforms like IP India Online, WIPO’s PatentScope, and the EPO’s Espacenet. Each database displays kind codes, and understanding them is essential to correctly interpreting what a search result actually represents – particularly in sectors like pharmaceuticals, where the distinction between a pending application and a granted patent has enormous commercial and public health implications.

How to read and use kind codes in practice 🔗

Kind codes used for patent documents published outside the U.S. vary by country – so when examining a foreign patent, always check the kind code table for that specific jurisdiction to ensure you have the correct document type and publication stage. The WIPO Handbook on Industrial Property Information and Documentation provides a comprehensive list of examples and kinds of patent documents for countries worldwide.

A few practical steps for working with kind codes: First, identify the country code (the two-letter prefix like US, EP, WO, IN). Second, locate the kind code at the end of the document number. Third, match it to the issuing office’s kind code table – available on the office’s website or the WIPO Handbook. Fourth, confirm the publication date, because some offices have changed their kind code systems over time and the same code may have meant something different before a certain year. The USPTO’s MPEP Section 901.04(a) is a reliable reference for U.S.-specific kind codes across different time periods.

Patent databases like Espacenet, PatentScope, and Derwent Innovation typically display kind codes prominently and often include hover-text or help sections explaining what each code means in context. Using these built-in tools can save significant time during a large-scale patent search.

What do you think? If two patent documents share the same number but carry different kind codes – say an A1 and a B2 – how would you approach using them together to assess the legal strength of a patent claim? And considering that India’s pharmaceutical sector is deeply affected by patent grants versus pending applications, how important do you think it is for IP professionals here to master kind code interpretation across multiple jurisdictions?

References

1. https://www.wipo.int/en/web/standards
2. https://ipindiaonline.gov.in
3. https://www.bitlaw.com/source/mpep/901-04-a.html