Introduction

Patentability searching sounds complex – faced with hundreds of millions of patent documents worldwide and a vast sea of scientific papers, where do you start? What are the steps? How do you ensure nothing is missed? This article breaks down the patentability search process into five actionable key steps, providing you with a complete roadmap from “receiving a technical disclosure” to “producing a patentability search report.”

Whether you are an IP newcomer conducting your first patentability search or a seasoned professional refining an existing workflow, mastering this five-step methodology will help you search more accurately, more comprehensively, and more efficiently. A clear patentability search process also makes it easier to document decisions and repeat the workflow across inventions.

Overview: The Five-Step Patentability Search Process

A complete patentability search process follows a structured, iterative process:

1. Understand the Invention — thoroughly grasp the technical solution, distill the inventive points
2. Develop a Search Strategy — construct a combined approach of keywords, classification numbers, and search queries
3. Execute the Search — systematically search across multiple databases
4. Comparative Analysis — compare the search results against the technical features of the invention on a feature-by-feature basis
5. Prepare the Report — deliver patentability search conclusions and recommendations

These five steps are not strictly linear — there are often feedback loops between steps. For example, discovering new technical terminology during the search may require returning to Step One to re-understand the invention; identifying a new search direction during comparative analysis may require returning to Step Three for a supplementary search.

Step One: Understand the Invention — The Foundation of Patentability Searching

Why This Step Is Most Easily Overlooked Yet Most Critical

The root cause of many failed patentability searches lies not in insufficient search skills, but in not truly understanding what the invention is about. If the understanding of the invention is inaccurate, the subsequent search direction will be off track, and the entire patentability search becomes “working efficiently” in the wrong direction.

How to Systematically Understand an Invention

Step 1: Read the Technical Disclosure

If the inventor has provided a technical disclosure document, you should understand it from the following dimensions:

• The technical problem to be solved: What real-world problem is this invention intended to solve? Why are existing solutions inadequate?
• The technical solution: What means does the invention employ to solve the problem? List all key technical features
• The technical effect: What improvements does it deliver compared to existing solutions? (faster speed, higher precision, lower cost, etc.)
• Specific embodiments: Are there concrete examples that help you understand the operational details of the solution?

Step 2: Communicate with the Inventor (if possible)

A technical disclosure often cannot fully convey all of the inventor’s ideas. When communicating directly with the inventor, consider asking the following questions:

• “What do you believe is the most fundamental difference between this invention and solutions already on the market?”
• “If you were to summarize the inventive point in a single sentence, how would you put it?”
• “Is there any technical background that you consider common knowledge but an outsider would not know?”

Step 3: Distill a Technical Feature Checklist

Decompose the invention into a set of specific technical features, each of which should be a searchable and comparable unit:

This checklist will play a central role in the subsequent “Comparative Analysis” step (Step Four).

Mistakes to avoid:

• Looking only at the title without examining the details, leading to a superficial understanding of key features
• Treating the inventor’s “subjective impressions” as technical features (e.g., “better user experience” is difficult to search)
• Overlooking technical features that appear secondary but are in fact critical to the invention

Step Two: Develop a Search Strategy — Design Your Search Plan

Keyword Strategy

The core challenge of patentability search is that the same technical concept may be expressed using entirely different terminology in different patents. Therefore, it is essential to prepare multiple synonymous expressions for each core technical concept:

Technical terms (the language used by R&D personnel):

• Example: if the invention involves “散热” (heat dissipation), search simultaneously for “heat dissipation,” “热管理” (thermal management), “冷却” (cooling), “cooling,” “thermal management,” “heat sink”

Functional terms (described from a functional perspective):

• Example: the “heat dissipation” function above can also be expressed in functional terms as “temperature control,” “温度控制” (temperature control), “thermal regulation”

Broader and narrower concepts:

• If your invention uses a “graphene heat dissipation film,” do not forget to search the broader concept “carbon material heat dissipation”; conversely, if your invention is a broad concept, search for narrower specific implementation materials

Chinese, English, and target market languages:

• A global patentability search must not search only Chinese. Chinese and English expressions of the same concept must be separately included in the search queries

Classification Number Strategy

Keyword searching has inherent limitations: different patents may describe the same technology using different terms. Classification number searching can compensate for this deficiency. In a reliable patentability search process, classification searching helps reduce the risk of missing documents that use unexpected terminology. Key classification systems include:

• IPC (International Patent Classification): The most fundamental global classification system, suitable for cross-border searching
• CPC (Cooperative Patent Classification): More granular than IPC, offering higher search precision
• FI/F-Term (specific to Japan): Must not be overlooked if the target market includes Japan

Practical recommendations:

1. First run a rough keyword search to find a few of the most relevant patents
2. Review the classification numbers of these patents
3. Analyze the definitions and hierarchy of the classification numbers to confirm they cover your technical field
4. Incorporate these classification numbers into the formal search query

Constructing the Search Query

Combine keywords and classification numbers into logical search queries. Commonly used operators:

• AND: Both conditions must be satisfied simultaneously (narrows the scope)
• OR: Satisfying either condition is sufficient (broadens the scope, typically used between synonyms)
• NOT: Exclude a specific condition (removes noise)
• Truncation / Wildcard (, ?): e.g., “cool” can match cool, cooling, cooled, cooler

Example search query:

(散热 OR 导热 OR heat dissipation OR thermal) AND (石墨烯 OR graphene OR 碳材料) AND IPC=H01L

Documenting the Search Strategy

Before the formal search, record the following information (this has important evidentiary value in subsequent examination proceedings):

• Search date and searcher
• List of databases used
• List of keywords and rationale for selection
• List of classification numbers and rationale for selection
• Search time range
• Complete search queries and hit counts

Step Three: Execute the Search — Systematic Multi-Database Searching

Database Selection Recommendations

Search Execution Recommendations

1. Start with the primary database: Typically begin with a global preliminary search on Espacenet or WIPO PATENTSCOPE
2. Iteratively refine the search query: If the search returns too many results (>1000), add restrictive conditions; if too few (<50), relax conditions or check whether the keywords are reasonable
3. Pay attention to the search time range: Ensure coverage from the earliest technical origins to the most recent published literature
4. Save all search records: Including the search queries, hit counts, and every step of the screening process

Step Four: Comparative Analysis — Feature-by-Feature Comparison Against Prior Art

This is the most intellectually demanding stage of patentability search work. You need to compare the prior art documents selected from the search against the technical features of the invention on a one-to-one basis. This is also where the patentability search process turns search results into practical filing guidance.

Comparative Analysis Methodology

1. Select the most relevant prior art documents: Typically 3–10 documents, noting the publication date and technical key points of each
2. Build a feature comparison table: Using the technical feature checklist extracted in Step One as the baseline, fill in row by row whether each prior art document discloses that feature
3. Novelty assessment: Does any single prior art document disclose all the technical features of the invention? If yes → a novelty issue exists
4. Inventive step assessment (if novelty is established): Identify the closest prior art → identify distinguishing technical features → determine whether a person skilled in the art would have been motivated, based on other prior art documents / common general knowledge, to arrive at the invention

Example Feature Comparison Table

Step Five: Prepare the Report — Conclusions and Recommendations

Elements a Patentability Search Report Should Contain

A professional patentability search report should at minimum contain:

1. Search information summary: Search date, searcher, databases, search queries, total hit count, and final number analyzed
2. Invention summary: Description of the technical solution based on Step One
3. Search results overview: List of key prior art documents selected (including publication number, publication date, title, patentee)
4. Feature comparison analysis: The comparison table from Step Four and explanatory analysis
5. Novelty conclusion: Single document comparison against each key prior art document, providing a novelty assessment
6. Inventive step conclusion: Analysis based on the problem-solution approach of whether “closest prior art + common general knowledge / other prior art documents” renders the invention obvious
7. Overall assessment: Comprehensive judgment of grant prospects (high / medium / low)
8. Recommendations:

• If grant prospects are favorable → recommend filing directly, indicate suggested claim strategy
• If risks exist → recommend amending the solution, supplementing experimental data, narrowing the scope of protection, etc.
• If grant prospects are low → recommend abandoning the application or switching to trade secret protection

Feedback Loops: Patentability Searching Is Not Linear

In actual patentability searching, feedback loops between steps are the norm:

• Discovering new technical synonyms during the search → Return to Step Two to supplement keywords
• Discovering a new relevant technical field during comparative analysis → Return to Step Three to supplement the search in that direction
• Feeling uncertain about a particular judgment while drafting the report → Return to Step Four to conduct additional comparisons or return to Step One to re-understand the invention

A good patentability searcher continuously optimizes through these loops until sufficient confidence in the conclusions is achieved.

Real-World Case Study: Applying the Five-Step Methodology

Scenario: A startup company has developed a foldable bicycle helmet.

Step One — Understand the Invention:

• Technical problem: Traditional helmets are bulky and inconvenient to carry
• Technical solution: The helmet consists of multiple arc-shaped shell segments connected by hinges; it can be expanded into a helmet form and folded into a compact form
• Key features: (1) multiple arc-shaped shell segments (2) hinge connection structure (3) volume reduction of more than 50% after folding

Step Two — Develop a Search Strategy:

• Keywords: 折叠头盔 (folding helmet), 可折叠头盔 (foldable helmet), collapsible helmet, foldable helmet, portable helmet
• Classification numbers: A42B 3/00 (helmets), A42B 3/32 (foldable helmets)

Step Three — Execute the Search:

• Search on Espacenet returned 85 hits; preliminary screening identified 12 relevant documents
• Supplementary searches on Google Patents and USPTO added 3 U.S. patents

Step Four — Comparative Analysis:

• Found 2 prior art documents relating to foldable helmets, but with different folding mechanisms
• Preliminary novelty assessment established (no single document disclosed all features)
• However, inventive step was questionable: D1 disclosed a hinge-connected folding structure, and D2 disclosed multiple arc-shaped shell segments for a helmet. Would a person skilled in the art have had the motivation to combine them?

Step Five — Prepare the Report:

• Conclusion: High probability that novelty is established; moderate risk for inventive step
• Recommendation: In the claims, focus on highlighting the specific combination of “arc-shaped shell segments + hinge” and the folding efficiency advantage it brings, rather than merely describing the broad concept of a “foldable helmet”

Actual outcome: The invention ultimately received a Chinese invention patent grant, but the scope of the granted claims was noticeably narrower than the original design.

Key Takeaway: The patentability search process follows five core steps – Understand the Invention (thoroughly grasp the technical solution), Develop a Search Strategy (keywords + classification numbers + search queries), Execute the Search (systematic multi-database searching), Comparative Analysis (feature-by-feature comparison against prior art), and Prepare the Report (conclusions and recommendations). Feedback loops exist between the five steps; good patentability searching is an iterative process of progressively approaching the truth.