International Patent Classification (IPC): The Global System for Organizing Patent Information

Every patent filed anywhere in the world needs to be findable – not just by the office that granted it, but by examiners, researchers, and businesses across the globe. That’s a formidable challenge when you consider that over 3 million patent documents are published each year. The International Patent Classification (IPC) exists to solve exactly this problem. It is a universal, language-independent system that organises every patent document by its underlying technology, making the world’s patent knowledge searchable, comparable, and analytically powerful.

The legal foundation: the Strasbourg Agreement of 1971 🔗

The IPC was established by the Strasbourg Agreement, signed on 24 March 1971, and administered by the World Intellectual Property Organization (WIPO). The Agreement created a formal treaty structure for how patents should be classified globally. It set up a Committee of Experts, consisting of representatives from all contracting states, to continuously revise and improve the system. Although only 68 states are formal parties to the Agreement, the IPC is used by the patent offices of more than 100 countries, four regional offices, and WIPO itself in administering the Patent Cooperation Treaty (PCT). The Strasbourg Agreement entered into force in 1975, at which point WIPO took sole responsibility for administering the classification. India, as a member of the Paris Convention, participates in this global framework.

What the IPC does: organising technology for retrieval 🔗

At its core, the IPC is a retrieval tool. It provides a hierarchical system of language-independent symbols for classifying patents and utility models according to the areas of technology they concern. When a patent document is published, an examiner at the national or regional patent office assigns one or more IPC symbols to it. These symbols appear on the front page of the published document and are indexed in databases, allowing anyone – from a student doing prior art research to a corporate IP analyst – to locate documents in a specific technical field without relying on keyword searches alone.

The language-independent nature of the IPC is particularly significant. Since classification is based on technical content rather than textual description, a researcher in India can use IPC codes to retrieve relevant patents filed in Japan or Germany without understanding those languages. This makes the system genuinely global in its utility.

The hierarchical structure of the IPC 🔗

The IPC is organised in a five-tier hierarchy, moving from the broadest to the most specific level of technical description. Symbols are arranged in a hierarchical, tree-like structure, from the highest level (sections) down through classes, subclasses, groups, and subgroups. Understanding this structure is essential for reading or searching any IPC code.

Sections: the eight top-level divisions 🔗

At the highest level, the IPC divides all technology into eight sections, each designated by a capital letter from A to H:

• A – Human necessities: food, personal hygiene, health, entertainment
• B – Performing operations; transporting: separating, mixing, shaping, printing, vehicles
• C – Chemistry; metallurgy: inorganic/organic chemistry, pharmaceuticals, materials science
• D – Textiles; paper: fibres, weaving, treatment of textile materials
• E – Fixed constructions: building, mining, civil engineering
• F – Mechanical engineering; lighting; heating; weapons; blasting: engines, pumps, refrigeration
• G – Physics: instruments, optics, computing, information storage
• H – Electricity: electric circuits, electronic devices, communications

Classes, subclasses, groups, and subgroups 🔗

Each section is broken down further. The eighth edition of the IPC – still the structural basis of the current system – comprises 8 sections, 129 classes, 639 subclasses, 7,314 main groups, and 61,397 subgroups. Across all levels, the IPC currently consists of over 74,000 classification codes.

A class symbol combines the section letter with a two-digit number – for example, C21 refers to metallurgy of iron within Section C. A subclass adds one or more letters – C21D covers modification of the physical structure of ferrous metals. A main group is expressed as a subclass symbol followed by a number and a slash, such as C21D 1/00, while subgroups subdivide main groups further, expressed with numbers after the slash, like C21D 1/26. The number of dots preceding a subgroup title in the printed classification indicates its hierarchical depth – more dots means more specific.

A complete IPC symbol therefore tells you a great deal at a glance. Take A61K 31/00: Section A (human necessities) → Class A61 (medical or veterinary science) → Subclass A61K (preparations for medical, dental, or toilet purposes) → Main group 31/00 (preparations containing organic active ingredients). A patent classified here is almost certainly a pharmaceutical compound patent.

Core level and advanced level 🔗

From the eighth edition onwards, the IPC introduced a two-tier system. The core level is intended for general information and searching smaller national collections, while the advanced level is intended for searching larger, international patent collections. Patent offices are required to classify documents at least to the core level. The advanced level, with its finer subgroups, provides the granularity needed for detailed prior art searches in large international databases like WIPO’s PATENTSCOPE or the European Patent Office’s Espacenet. This distinction is important for practitioners: when doing a quick national search through India’s InPASS portal, core-level classification is generally sufficient; for a global freedom-to-operate analysis, the advanced level is necessary.

How IPC symbols are assigned in practice 🔗

Before a patent document is published, an examiner at the relevant patent office classifies it by assigning the IPC symbol or symbols corresponding to the technical field of the invention. For PCT applications filed under the international patent system, IPC symbols are allotted by the International Searching Authority (ISA) – the designated authority that conducts the international search. The symbols are then printed on the front page of the published application.

In India, the Indian Patent Office (IPO) assigns IPC codes to all patent applications during examination. These codes appear on published patent documents and are searchable via the Indian Patent Advanced Search System (InPASS), which allows users to filter results by IPC. For convention filings (applications claiming priority from a foreign application), Indian Form 1 explicitly asks for the IPC classification used in the convention country. For PCT national phase entries, the international publication already carries IPC symbols, which flow directly into the Indian record.

Annual revision: keeping pace with new technology 🔗

Technology evolves constantly, and a classification system frozen in 1971 would rapidly become obsolete. The IPC is continuously revised, with a new version entering into force each year on January 1. This revision is carried out by the Committee of Experts. New subgroups are added when emerging technologies – such as artificial intelligence, nanotechnology, or gene editing – generate enough patent activity to warrant their own classification codes. Older subgroups may be merged or restructured. For the first seven editions, updates happened roughly every five years; the shift to annual revisions from the eighth edition onwards reflects the accelerating pace of technological change.

For students and practitioners, this means the IPC version cited in a patent document matters. When reading classification codes in databases, the year and IPC edition appear in brackets, and documents may not yet be reclassified under the newest edition, so a potential lag time should be kept in mind.

The IPC as a tool for patent analytics and technology intelligence 🔗

Beyond search and retrieval, the IPC has become a foundational instrument for patent analytics – the systematic analysis of patent data to extract strategic and policy insights. Because every patent document carries at least one IPC code, the classification enables structured, large-scale analysis that keyword searches cannot match.

Identifying technological trends 🔗

The IPC allows trends in particular areas of science and technology to be identified for statistical purposes, networks of co-occurrences of classification codes to be mapped, and – combined with citation analysis – the main trajectories of emerging fields to be traced. Governments, research institutions, and corporations use IPC-based analytics to understand where innovation is concentrated, where it is declining, and where white spaces for new investment exist.

The IPC Green Inventory 🔗

One of the most practically significant applications of the IPC in recent years is the IPC Green Inventory, a specialised tool launched by WIPO in 2010 that contains around 200 topics directly relevant to environmentally sound technologies (ESTs), each linked to the most relevant IPC symbols chosen by experts from around the world. The inventory is hyperlinked to WIPO’s PATENTSCOPE database, allowing researchers to instantly identify international patent applications in green technology areas. It covers fields including alternative energy production, energy conservation, transportation, waste management, and agriculture and forestry. For Indian researchers, businesses, and policymakers working on climate and sustainability goals, the IPC Green Inventory provides a direct route to understanding the global landscape of green innovation.

IPC vs. Cooperative Patent Classification (CPC) 🔗

Practitioners will frequently encounter a related system: the Cooperative Patent Classification (CPC), jointly operated by the European Patent Office (EPO) and the United States Patent and Trademark Office (USPTO). The CPC is built as a more detailed extension of the IPC, using just over 260,000 classification codes compared to the IPC’s 74,000. The CPC also includes a Section Y for cross-sectional, emerging technologies – a category absent from the IPC. In practice, Indian patent documents display IPC classification, while European and US documents typically show both IPC and CPC. For a thorough international search, starting with IPC to frame the technology area and then using CPC for deeper precision – especially for prior art in US and European databases – is the most effective approach.

Practical relevance for Indian patent professionals and students 🔗

For law students, patent agents, and IP professionals in India, the IPC is not an abstract classification exercise – it shapes daily practice. Assigning an IPC symbol does not expand or restrict the legal scope of patent claims; it is an indexing label, not a rights conferrer. However, an incorrect or imprecise classification may mean that relevant prior art is missed during examination, which can indirectly affect the quality and defensibility of a granted patent.

Proficiency with the IPC also opens up strategic possibilities. Monitoring IPC subclasses in a competitor’s technology domain reveals their innovation direction. Analysing the volume of filings in a particular subgroup over time can signal whether a technology is maturing or still in rapid development. For startups and research-intensive firms in sectors like pharmaceuticals (Section A61), electronics (Section H), or clean energy (intersecting Sections F and C), IPC-based analytics can directly inform R&D investment decisions.

WIPO provides several supporting tools free of charge: IPCCAT (a computer-assisted categorization tool to help classify patents at IPC class, subclass, or group level), and STATS (which generates IPC predictions based on statistical analysis of patent documents). The Guide to the IPC, the official document approved by the bodies of the Strasbourg Agreement, explains classification principles with worked examples and is freely available on WIPO’s website.

What do you think? As India continues to expand its patent filing activity, should Indian universities integrate IPC-based patent analytics into their science and engineering curricula alongside law schools? And given that IPC annual revisions now incorporate technologies like AI and biotechnology, how should patent examiners and attorneys continuously update their classification expertise to keep pace?

References

1. https://www.wipo.int/en/web/treaties/classification/strasbourg/summary_strasbourg
2. https://en.wikipedia.org/wiki/International_Patent_Classification
3. https://www.wipo.int/en/web/classification-ipc/preface
4. https://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:International_patent_classification_(IPC)
5. https://www.dpma.de/english/search/classifications/patents_and_utility_models/ipc/index.html
6. https://wipo-analytics.github.io/handbook/classification.html
7. https://www.wipo.int/edocs/pubdocs/en/wipo_guide_ipc_2009.pdf
8. https://abhijitbhand.com/blog.php?slug=what-are-ipc-and-cpc-classifications-in-patents&v=1758948705
9. https://www.wipo.int/pressroom/en/articles/2010/article_0031.html
10. https://www.sciencedirect.com/science/article/pii/S0959652624038344
11. https://www.wipo.int/en/web/classification-ipc