Introduction

Patentability search pitfalls often appear after a report seems complete. The patentability search report says “favorable prospects for grant,” and you submit the application with full confidence — but a few months later, the examiner cites two prior art documents and demolishes both novelty and inventive step. You dig out the patentability search report, only to find that neither of those two documents appeared in the search results.

This is not the examiner “out to get you”; this is a classic case of — an incorrect patentability search conclusion. Not intentional, but wrong is wrong.

This article dissects the five most common patentability search pitfalls and the countermeasures for each. Know where you’re likely to stumble, and you’ll walk more steadily.

Patentability Search Pitfalls 1: The Keyword Blind Spot — “No One Else Uses the Words I Searched”

The Core Problem

You search for “mobile phone heat dissipation structure,” but the prior art document uses “thermal management assembly for portable electronic devices” — it describes the same thing, but your keywords miss it entirely.

Why This Pitfall Is Especially Dangerous

It gives you a false sense of security that “no prior art was found.” You believe you’ve “searched thoroughly,” when in reality you’ve only confirmed that “no one uses the words I searched.”

Typical Manifestations

• The search yields no more than 20 results, and you think, “This field really is novel after all.”
• In truth, there may be 200 relevant patents — your keywords simply didn’t cover them.

Countermeasures

1. Build a keyword synonym matrix for every core concept (see the upcoming article “Keyword Thinking Tools” for details)
2. Classification-based search as a safety net: regardless of what terminology others use, if the technology falls under the same classification, it won’t slip through
3. Find one known relevant patent, examine the vocabulary it uses, and incorporate those terms into your search strategy
4. After completing your search, reverse-verify against several patents you “know for certain” are relevant: can your search strategy retrieve them? If not → your search coverage is insufficient

Pitfall 2: Hindsight Bias — “It’s Obvious Once You Know the Answer”

The Core Problem

You already know every detail of the present invention, and then you look back at the prior art — everything seems to scream, “Isn’t this plainly obvious?”

This is an inherent limitation of human cognition: once you know a solution, it is extremely difficult to reconstruct how you would have thought without knowing that solution.

Why This Pitfall Is Especially Dangerous

It causes you to be excessively harsh in inventive step assessment, potentially “self-rejecting” an invention that genuinely possesses inventiveness — costing the enterprise an innovation that should have been granted a patent.

Typical Manifestations

• “This invention is merely combining A with B — I could have done that too.”
• But from the perspective of a person skilled in the art as of the filing date, selecting exactly the A+B combination from hundreds or thousands of possible combinations and achieving a specific technical effect may well not be obvious.

Countermeasures

1. Explicitly flag “hindsight bias checkpoints”: in the conclusion section of the patentability search report, proactively ask: “Has my judgment been influenced by the fact that I already know the invention?”
2. Search for evidence that “no one thought to do it this way”: if the technical problem existed for many years, and numerous patents attempted similar approaches but none moved in this direction — that constitutes positive evidence of inventiveness
3. Interview the inventor about the R&D journey: what approaches did they try? Why didn’t they succeed? What did other practitioners do? — this helps you reconstruct the genuine technical judgment as of before the filing date

Patentability Search Pitfalls 3: Insufficient Search Scope — “The Fatal Art Just Happens to Be Where I Didn’t Search”

The Core Problem

You searched Chinese patents and found no prior art. But the fatal prior art document is a Japanese patent (with FI/F-Term classification) or a 2022 arXiv paper.

Why This Pitfall Is Especially Dangerous

It is the “silent killer” — you write “no relevant prior art identified” in the report, when the reality is not “it doesn’t exist” but “you don’t know it exists.”

Typical Manifestations

• “I searched Espacenet — isn’t that comprehensive enough?”
• In reality, Espacenet’s coverage of Chinese patents is incomplete, and its classification mapping for Japanese patents has limitations.

Countermeasures

1. Clearly state the scope “not covered” in the patentability search report, rather than writing a vague “comprehensive search”
2. Cross-validate across at least three databases: Espacenet + PATENTSCOPE + Google Patents is the baseline
3. If the technical field has obvious regional characteristics (e.g., precision machinery in Japan, semiconductors in South Korea), you must search the patent database of that country
4. Explicitly list the search scope in the report so the reader knows “what was not searched”: this is actually more responsible than “claiming to have searched everything”

Pitfall 4: Gross Underestimation of Non-Patent Literature

The Core Problem

See Article 9 for detailed discussion — 30–40% of prior art exists only in NPL. In fields such as AI and biotechnology, this proportion is even higher.

Countermeasures

• Enforce as a mandatory step in the patentability search SOP: every invention patentability search must include at minimum a Google Scholar search
• For inventions involving software / AI, supplement with arXiv and GitHub searches
• Clearly state in the report: whether NPL was searched, and which platforms were searched

Pitfall 5: Subjective Fluctuation in the Inventive Step Standard

The Core Problem

The same invention receives radically different inventive step assessments from different patentability search analysts — some consider it “obvious,” others regard it as possessing “inventive height.” This fluctuation arises from differing assumptions about the capabilities of the person skilled in the art and differing criteria for judging “technical teaching.”

Countermeasures

1. Introduce an “examiner-perspective review”: have another patentability search analyst or patent attorney independently review your inventive step assessment
2. Conduct more hypothetical reasoning: “If I were the examiner, what prior art documents would I use to challenge this? What would my chain of challenge look like?”
3. Focus on the “refutability” of the conclusion: when your inventive step assessment faces an examiner’s challenge, how much room is there to rebut?

Pitfall Checklist

After every patentability search, self-check with the following questions:

Regarding Search:

• [ ] Can my search strategy retrieve known relevant patents? (reverse verification)
• [ ] Did I search at least three databases?
• [ ] Did I conduct an NPL search?
• [ ] If the technology originates from countries such as Japan or South Korea, did I search the local databases?

Regarding Assessment:

• [ ] Is my assessment affected by hindsight bias?
• [ ] Are my assumptions about the capabilities of the person skilled in the art reasonable?
• [ ] Has someone independently reviewed my assessment?

Regarding the Report:

• [ ] Does the report clearly state the limitations of the search?
• [ ] Does the report indicate which NPL sources were searched and which were not yet searched?

Key Takeaway: Patentability search pitfalls can distort both search coverage and legal assessment. The five pitfalls in patentability search — (1) the keyword blind spot leading to omissions, (2) hindsight bias leading to excessive rejection of inventive step, (3) insufficient search scope leading to missed prior art, (4) gross underestimation of NPL, and (5) subjective fluctuation in the inventive step standard. Each pitfall has clear identification signals and countermeasures. Self-checking with the checklist after completing a patentability search can effectively reduce these risks.

To reduce recurring patentability search pitfalls across a team, PatSnap Analytics can help standardize search records, classification checks, and prior art review workflows.