HINDSIGHT BIAS IN THE ASSESSMENT OF INVENTIVE STEP
Learn what hindsight bias is in patent law, why it affects inventive step analysis, and how India, Europe, the United States, and WIPO prevent ex post facto reasoning.
By Sakthivel
In patent prosecution, hindsight bias (also called ex post facto analysis) arises when an examiner or opponent reads the invention, sees the solution, and then works backward through prior art to reconstruct a path that makes the invention appear obvious. Because the solution is already known, the reconstruction often feels logical, even though it would not have been apparent to a person skilled in the art at the time of filing. This is a dangerous scenario for inventive step analysis because the examiner or court subconsciously uses the invention as a roadmap, making it look easier than it truly was. Legal systems therefore insist that inventive step must be assessed from the viewpoint of a notional “person skilled in the art” at the priority date, without knowledge of the invention’s solution or later commercial success.
When reviewing how different jurisdictions guard against hindsight in inventive step analysis, a clear pattern emerges as each system has developed structured safeguards to ensure that the invention is judged from the perspective of the skilled person at the priority date, not reconstructed with knowledge of the solution.
In India, the Patent manual emphasizes that an Invention must be considered as a whole. In other words, it is not sufficient to conclude that a claimed invention is obvious merely because individual parts of the claims, taken separately, are known or might be found to be obvious. Furthermore, in Salmond L. J. in Rado v. John Tye & Son Ltd., suggested another test for determining whether a document would negate novelty or inventive step: Had the document been placed in the hands of a competent craftsman (or engineer, as distinguished from a mere artisan), endowed with the common general knowledge at the priority date, and faced with the problem solved by the patentee but without knowledge of the patented invention, would that craftsman have arrived at the claimed solution?
In European practice, the Boards of Appeal have developed the Problem–Solution Approach (PSA) as a strict safeguard against hindsight. The examiner must begin by identifying the closest prior art objectively, selecting a single piece of prior art that represents the most promising starting point for the skilled person. From there, the objective technical problem is defined by comparing the claimed invention with that prior art, identifying the distinguishing features, and determining the technical effect those features achieve. The final step is the “could-would” analysis, which asks whether the skilled person would have been motivated to modify the closest prior art to solve the objective technical problem, not simply whether they could have done so.
The Boards of Appeal repeatedly emphasize that just because all the necessary information existed in the prior art does not mean the skilled person would have combined it. There must be a clear teaching, hint, or motivation in the prior art itself. The closest prior art cannot be chosen by working backward from the invention, but must be selected based on subject matter conceived for the same purpose or aiming at the same effect. If a combination requires a convoluted set of sequential steps that only become apparent when starting from the claimed invention and working backward, the Board will reject it as hindsight.
Case Law of the Board of Appeal of the European Patent Office consistently warns against “ex post facto” analysis, the EPO’s term for hindsight. Obviousness must not be based on knowledge of the invention, and the invention cannot be reconstructed from prior art using information available only from the patent itself. The Guidelines and decisions underline three key points: inventive step must be assessed without knowledge of the invention as claimed; the choice of closest prior art, the definition of the objective technical problem, and the combination of documents must be made without building in the inventor’s solution; and the PSA is expressly described as a tool to reach decisions without hindsight. For example, the objective technical problem is derived only from the differences and technical effects over the closest prior art, not from the inventor’s subjective problem statement, precisely to reduce the risk of hindsight.
Get Altacit Global’s stories in your inbox
Join Medium for free to get updates from this writer.
This structured approach ensures that inventive step analysis in Europe remains forward, looking, grounded in the realities of the field at the priority date, and insulated from the temptation to reconstruct the invention with knowledge that only hindsight provides.
In USA, the Graham v. John Deere framework is the landmark United States Supreme Court test established in 1966 (383 U.S. 1) to determine whether an invention is “non-obvious” and therefore eligible for a patent under 35 U.S.C. § 103. Courts, attorneys, and the USPTO apply this framework sequentially to avoid hindsight bias.
Firstly, the scope and content of the prior art must be defined. This involves identifying what knowledge was publicly available: patents, publications, and other disclosures that a person in the field had access to before the invention’s filing date.
Secondly, the differences between the prior art and the claims must be identified. This step pinpoints the exact technical distinctions between what already existed and the new elements being claimed.
Thirdly, the level of ordinary skill in the art (PHOSITA) must be determined. This requires assessing the education, experience, and knowledge base of a hypothetical skilled person in the relevant technical field, since inventive step is always judged from their perspective.
To further guard against hindsight, the framework also requires consideration of secondary factors, sometimes called objective indicia of non-obviousness. These include commercial success, which shows the invention solved a problem and achieved market acceptance; long-felt but unresolved need, where the industry struggled for years until the invention provided a solution; failure of others, which demonstrates that competitors attempted but could not achieve the same result; and unexpected results, where the invention performs noticeably better or differently than prior art would suggest.
By combining these factual inquiries with objective evidence, the Graham framework ensures that obviousness is assessed fairly, without reconstructing the invention from hindsight, and always from the viewpoint of the skilled person at the time the invention was made.
The WIPO study makes it clear that combining prior art is especially vulnerable to hindsight, and it provides a structured checklist to test whether such a combination is realistic. Firstly, you must look at the nature and content of the teachings: if the documents come from different subfields, serve different purposes, or even contradict each other, that weakens the case for obviousness. Secondly, consider the number of selections or pieces to combine: the more cherry-picking required, the stronger the hindsight argument, since the examiner is effectively building a mosaic. Thirdly, check whether the prior art teaches away: if it discourages the claimed modification, that is strong evidence against obviousness. Fourthly, examine technical trends and demands: if the field was moving in a different direction at the relevant time, the claimed path is less likely to have been obvious. Fifthly, assess the age of the documents: very old prior art combined with a long-felt need supports non-obviousness, because if the solution were obvious, someone would likely have implemented it earlier. Sixthly, highlight any advantageous or surprising effects: unexpected or synergistic results show that the combination goes beyond routine development. Finally, seventhly, evaluate predictability: if the result is not just a predictable outcome of routine optimization, that strengthens the case against hindsight.
When framed in argument, these factors allow you to demonstrate why a proposed combination is unrealistic. For example: “Combining D1 and D2 requires multiple selections across unrelated teachings, and neither document suggests applying feature F from D2 to the context of D1. WIPO’s guidance recognizes that such reconstructions are especially susceptible to hindsight, particularly when the field trend was towards [different approach] and when the claimed combination yields an unexpected effect of [effect].
By anchoring objections in WIPO’s checklist, you move beyond simply asserting hindsight and instead demonstrate, point by point, why the examiner’s reconstruction relies on knowledge of the invention rather than the perspective of the skilled person at the priority date. This structured approach makes hindsight arguments far more persuasive and aligned with international best practice.
In view of the above, patent manuals and case law agree on one core idea: inventive step must be judged from the perspective of the skilled person before the invention was known. The best rebuttals do not merely say “this is hindsight”; they show exactly where the examiner has used the invention as a roadmap and then anchor the answer in the patent manual’s own framework. That is usually the strongest way to defeat an obviousness/inventive step objection based on hindsight.
