A patent specialist receives a short letter from an engineer: “we need to check whether our development is patentable,” and is forced to guess for themselves what exactly to research, in which countries to search, to what depth, and in which IPC classes. The result of such a task setting is predictable: the search turns out either too broad and takes weeks without a specific conclusion, or too narrow and misses key analogues. The problem is not the qualification of the patent specialist, but that the assignment for conducting patent research was never formulated as a technical task with clear boundaries in the first place.
In this article we will examine which tasks need to be defined before the search begins, who takes part in organizing the research, how to choose the object and subject of analysis, and what should be included in the finished assignment before the research is launched.
Patent research tasks: what must be defined before the search begins
Before formulating a specific assignment, it is important to understand for what practical purpose the research is being conducted, since the depth of the search, the set of sources, and the final structure of the report depend on it. The tasks of patent research usually fall into one of several categories: assessing the patentability of the solution being developed, checking freedom to operate (patent clearance) before bringing a product to market, analyzing the state of the art and development trends of the industry, and searching for potential partners or competitors based on patent activity.
Each of these tasks requires a different approach to the search. Checking patentability concentrates on a narrow subject area and requires the widest possible coverage of countries and the maximum search depth, because even one missed analogue can make an application unpatentable. Checking patent clearance, by contrast, is limited to the countries where the product is to be sold or manufactured, but requires careful analysis of precisely the patents in force, rather than the entire historical mass of documents. Mixing these tasks in one vaguely set assignment is a frequent reason why the result of the research does not answer the question for which it was commissioned.
Organization of patent research: who takes part and what data is needed
Organizing patent research involves at least two parties with different areas of responsibility: the engineer or developer, who knows the technical essence of the solution, and the patent specialist, who translates this essence into a formalized search query and interprets the results found. An assignment drawn up by only one of the parties without the involvement of the other usually turns out to be incomplete: the engineer does not always know what data the patent specialist needs for a high-quality search, and the patent specialist, without a deep immersion in the technical specifics, may miss the essential features of the solution.
To organize the research effectively, the following are needed from the engineer or customer at a minimum: a technical description of the solution indicating the essential features that distinguish it from known analogues; the field of application and the expected sales market; analogues or similar solutions known to the developer, even if they have not been systematically checked; the intended countries of patenting or sale of the product. The more complete these initial data, the more precisely the patent specialist will be able to formulate the search query and the fewer clarification iterations will be required during the work itself.
Fundamentals of patent research: how to define the object and subject of analysis
The fundamentals of patent research are built on distinguishing between two concepts that are often confused in practice: the object and the subject of analysis. The object of research is the specific technical solution, product, device, or technological process with respect to which the check is conducted. The subject of the search is those features of the object by which the comparison with patent documentation is made: design features, material composition, the way a function is implemented, technological parameters.
The difference between the object and the subject is fundamental for the quality of the result. If the assignment specifies only the object (“check the new heat exchanger”) without specifying the essential features, the patent specialist has to either interpret the task at their own discretion or go back to the customer for clarifications, losing time. If, however, the subject of the search is clearly specified in the assignment, that is, a specific design feature or technical principle that the team considers potentially new, the search immediately becomes targeted and makes it possible to obtain a result in fewer iterations.
Separately, the assignment should define the search depth, that is, the period for which patent documents are reviewed, and the search countries corresponding to the task of the research. For checking patentability, the search depth is usually maximal, since even a long-filed and already inactive application can indicate a lack of novelty. For checking patent clearance, what matters first of all are the patents in force in the specific countries where commercial use of the solution is planned.
Content and procedure of patent research: what to include in the assignment
The content and procedure of patent research are fixed in the assignment as a formalized document on the basis of which the patent specialist begins work. Such an assignment is usually prepared in accordance with GOST R 15.011-96 and includes: the name and purpose of the object of research, the purpose of the work, the object and subject of the search with an indication of the essential features, the countries and depth of the search, the list of sources of patent and scientific and technical information that must be covered, requirements for the structure of the final report, and the deadlines for the work.
It is important that the assignment records not only what needs to be found, but also what conclusions the final report must contain: a comparison of the object with the closest analogues, an assessment of patentability or patent clearance, and recommendations for adjusting the technical solution if the search revealed close analogues. Without this part of the assignment, the patent specialist may limit themselves to a simple list of the documents found, leaving the interpretation of the results to the customer’s discretion, which reduces the practical value of the entire research.
Stages of conducting patent research after the assignment is approved
After the assignment has been agreed and approved, the research goes through several sequential stages. First, the patent specialist develops the search regulation: specific search queries, International Patent Classification codes, and the list of databases that will be used. Then the search of patent documentation itself is carried out according to the approved regulation, covering the countries and depth specified in the assignment.
The documents found are analyzed and compared with the object of research according to the essential features specified in the assignment; at this stage a comparison table is prepared, which forms the basis of the conclusions. Next, the patent specialist prepares the final report, the structure of which must correspond to the requirements fixed in the assignment: analysis of the state of the art, a list of identified analogues, an assessment of patentability or patent clearance, and recommendations for further action.
The final stage is agreeing the report with the customer: the engineer or developer for whom the research was conducted checks whether the conclusions answer the original question posed in the assignment and, if necessary, requests clarification of individual points. If, following the agreement, it is necessary to expand the depth or countries of the search (for example, because the analogues found point to an adjacent field that was not taken into account in the original assignment), this is formalized as a supplement to the assignment, not as a reason to start the research over from scratch.
How “POSINT” helps formulate the assignment and conduct the research faster
Formulating a high-quality assignment requires time and experience: one must understand in advance which IPC codes are relevant to the topic, what search depth is sufficient for the specific task, and which countries should be covered. An error at this stage almost inevitably leads to the finished report having to be reworked or requested again.
The AI engineer “POSINT” simplifies precisely this part of the work:
- accepts a description of the object of research in natural language, with no need to formulate the query in advance in terms of patent classification;
- independently determines the relevant IPC codes and forms a search query based on the described essential features;
- analyzes patent documentation in Russian and foreign databases, covering the required search depth;
- compares the object of research with the analogues found and generates a structured report with an assessment of novelty.
This allows the engineer or patent specialist to obtain a full research result in hours instead of weeks of manual work, and, if necessary, to adjust the depth or direction of the search immediately, without waiting for a repeat iteration. Run a patent search with “POSINT” already at the task-setting stage: this will save time on agreeing the assignment and speed up receipt of a practically applicable report.
Conclusion
A high-quality patent research starts not with searching for documents, but with a correctly formulated assignment. The more precisely the purpose of the research, the object and subject of analysis, the essential technical features, the countries, the search depth, and the requirements for the final report are defined, the lower the risk of getting either an excessive array of irrelevant documents or an incomplete selection without key analogues. Therefore, setting the task is not a formality but one of the determining stages of the entire patent research.
“POSINT” helps reduce the time needed to prepare and conduct the search: the AI engineer analyzes the technical description, determines the relevant search directions and IPC codes, finds patent analogues, and compares them with the solution under study. This makes it possible to move faster from the initial technical idea to a structured result that can be used for making further engineering and patent decisions.
Do not spend weeks on manual searching for patent analogues and on agreeing the search assignment. Submit a request to “POSINT” and provide a description of your development: the AI engineer will help determine the direction of the research, find relevant patent documents, and assess your development relative to existing technical solutions.