Introduction

A patentability search decision tree helps IP teams choose the right search depth before spending time or budget. When facing a new invention, the patentability search strategy varies significantly depending on the circumstances. “Should this undergo a deep patentability search or a DIY check?” “Do I need to search Non-Patent Literature?” “How many databases should I search?”

To answer these questions, this article provides a patentability search decision tree and a Patentability Search Decision Tree — regardless of your invention’s type, target market, or budget, this decision tree helps you quickly identify the right patentability search route.

Patentability Search Decision Tree

Follow the decision flow below, evaluating each node layer by layer. In short: high-value invention patents usually need standard or multi-jurisdiction review; lower-risk utility models may start with expedited search or DIY self-check.

Route A: Standard Single-Market Patentability Search (Invention Patent / Core Innovation, Single Target Market)

Selection Criteria:

• Invention type: Invention Patent or core technology
• Target market: Single country or region
• Budget: Adequate

Execution Steps:

1. Understand the invention → Extract core features
2. Build search strategy → English keywords + target market local language keywords + at least 3 primary IPC classification numbers
3. Search at least 3 databases:

• Espacenet (global patent database, primary search tool)
• Target market patent office database (e.g., USPTO, CNIPA, JPO, etc.)
• Google Patents (supplementary and cross-validation)

1. NPL search: At minimum Google Scholar (research papers) + target market academic databases
2. Feature comparison → Problem-Solution Approach analysis → Report

Estimated Time Investment: 4–8 hours per case (by a professional patentability searcher)

Route B: multi-jurisdictional route (Core Invention, Multi-Country / Global Filing Strategy)

Selection Criteria:

• Invention type: Core platform technology, high-value invention
• Target market: Multiple countries (including the U.S., Europe, Japan, or more)
• Budget: Ample

Execution Steps:

1. Thoroughly understand the invention
2. Build a multilingual search strategy:

• Keywords: Prepare a set each in Chinese + English + the local language(s) of the target market(s)
• Classification numbers: IPC (baseline) + jurisdiction-specific classification supplements (Japan = FI/F-Term, U.S. = CPC deep subdivisions)

1. Search databases (at least 5):

• Espacenet (global database, primary search tool)
• USPTO (U.S. patents; mandatory for the U.S. market)
• Target market patent office databases (CNIPA, JPO, KIPO, etc., selected according to target markets)
• J-PlatPat (Japanese patents; relevant when the Japanese market is involved or Japan leads in the relevant technology field)
• Google Patents (supplementary and cross-validation)

1. NPL search:

• Google Scholar + target market academic databases (e.g., IEEE / ACS / PubMed, depending on the technical field)

1. Perform differentiated patentability assessment for each target market’s examination practice (Inventive Step standards differ across jurisdictions)
2. Full feature comparison → Problem-Solution Approach analysis → Multi-jurisdictional strategy recommendations

Estimated Time Investment: 10–20 hours per case

Route C: Expedited Patentability Search (Utility Model / Incremental Improvement / Minor Iteration)

Selection Criteria:

• Invention type: Utility Model / short-term patent or incremental improvement
• Importance: Non-core product, or temporary / peripheral protection
• Budget: Limited

Execution Steps:

1. Simplified understanding of the invention → Extract 2–3 core features
2. Rapid search strategy → Primary English keywords (a complete synonym matrix is not required)
3. Search 2 databases:

• Google Patents (primary search tool)
• Espacenet (supplementary verification)

1. NPL: Optional (a quick scan of Google Scholar suffices)
2. Quick assessment: Does any single reference “fully cover” all the features?

• Yes → Abandon; adjust the invention
• No → Favorable grant prospects (do not pursue an Inventive Step deep-dive — the Inventive Step requirement for Utility Models / short-term patents is inherently lower than that for standard Invention Patents)

Estimated Time Investment: 1–2 hours per case

Route D: DIY Self-Check (Non-Critical Utility Model / Initial Self-Screening by R&D Personnel)

Selection Criteria:

• The invention is a non-critical product
• No IP team intervention is needed at this stage
• Budget: Free / Zero

Execution Steps:

1. Write a brief description of the invention (1 paragraph)
2. Search on Google Patents using natural language (primarily in English; supplement with local language if necessary)
3. Browse the top 30 patents returned
4. If you see anything highly similar → Report to the IP team
5. If you see nothing highly similar → Positive grant expectation; proceed to engage the IP team

Estimated Time Investment: 30–60 minutes per case

Decision Tree Companion Assessment Rubric

After completing the patentability search and when confronting prior art references, use the table below to quickly determine the next step:

Decision Tree Usage Examples

Scenario 1:
“We are a new-energy battery startup. We have a core formulation innovation related to solid-state electrolytes and intend to file patents simultaneously in Europe and the United States.”

→ Path: Q1 → ‘Invention Patent / High-Value Core’, Q2 → ‘Multi-Country / Global’ → Execute Route B: multi-jurisdictional route. The U.S. route requires focused CPC classification numbers and may require supplementary searching on USPTO; EPO examination demands a strengthened “technical character” argument.

Scenario 2:
“I run a cross-border e-commerce business. I designed a foldable storage bin and want to file a Utility Model / short-term patent domestically first. If market feedback is positive later, I’ll consider multi-country Invention Patent filings.”

→ Path: Q1 → ‘Utility Model / Incremental Improvement’, Q2′ → ‘No’ (non-critical product) → Execute Route C: Expedited Patentability Search. If cost is a significant concern, you may drop down to Route D: DIY Self-Check. However, if the search reveals a highly competitive patent landscape → escalate to a professional patentability search.

Scenario 3:
“We are a medium-sized medical device manufacturer. We have developed a relatively novel ultrasound probe structure targeting the European and U.S. markets. It serves as the foundational technology for our flagship product over the next three years.”

→ Path: Q1 → ‘Invention Patent / High-Value Core’, Q2 → ‘Multi-Country / Global’ → Route B. Because it is a medical device involving biological effects, the NPL search should be further expanded to include PubMed and Google Scholar for clinical literature.

Key Takeaway: A patentability search decision tree helps match search depth to risk, market, and budget. The Patentability Search Decision Tree helps you quickly identify the correct patentability search route based on two core dimensions — (1) invention type (Invention Patent / High-Value Core vs. Utility Model / Incremental Improvement), and (2) target market (Single Market vs. Multi-Country / Global). The four routes, ranging from “Standard Single-Market Patentability Search” to “DIY Self-Check,” are tiered by risk level and budget gradient. Important note: the routes are not meant to be repeated — if, after executing a lower-tier route, you discover “red flags” during the search, you should promptly escalate to a higher-tier route and supplement with a professional patentability search.

For teams applying a patentability search decision tree across many disclosures, PatSnap Analytics can help compare patent landscapes, assignee activity, and prior art risk before routing cases.