One of the most precarious decisions to make during the patenting process is when to commence filing. There is an inherent tension between getting an early priority date which is potentially light on data, as opposed to accumulating plenty of supporting data but having a later than desirable priority date.

In some cases, for instance where there is an unintended or ill-advised early disclosure of the invention or a critical commercial development, the requirement for an early priority date overrides any consideration of the contents of the patent specification and the only course of action is to include whatever data is at hand. Fortunately, in most cases though, patent applicants have a degree of discretion about when to commence filing, in which case the balance between an early priority date and potential patent quality is an issue.

All jurisdictions have some sort of "enablement" requirement, meaning that the patent applicant needs to provide a description of how to put the invention into practice over the full scope of the claims. In general, the description must allow the skilled person to put the invention into practice without an undue burden of trial and experiment, or the need for further invention.

Ensuring solid enablement at filing is particularly important, as unlike other potential hidden defects in a patent specification that can surface during prosecution or enforcement, such as a lack of novelty or inventive step, problems with enablement are not readily corrected after filing.

Unfortunately, the answer to the question "How much experimental detail is enough" is the same as every question in patent law, which is "it depends". While there is no prescriptive formula for how much data is enough, there are a few guiding principles which can assist in arriving at a well-informed conclusion.


The first step in considering if the available information or data is enough is to try to consider the position from the point of person skilled in the art, that notional reader to which every patent specification is addressed.

Framing the notional person skilled in the art correctly is an important step because it sets a bar for the level of "assumed knowledge" that can be imputed into the specification and can be considered as the baseline disclosure. There is no need for this assumed knowledge to be explicitly set out in the description.

The level of disclosure is be determined by the subject matter and will vary between patent specifications depending on the likely audience of those people with a practical interest in carrying out the invention. The level of assumed knowledge can be very high in some technical fields if the skill of practitioners is high.

The person skilled in the art is a highly artificial construct, who is assumed to know everything about routine work in the field while simultaneously being entirely lacking in any inventive capacity. More confusingly, the skilled person may also be a team of people, again, each very knowledgeable in their field and capable of working together as one but having no ability to create or develop any new concepts.

In general, inventors who provide the details of their invention tend to have reasonably intuitive grasp of what is and what is not common general knowledge. What they assume is a given will probably be the same as their peers, but it is important to ensure that anything more than standard operating procedures is fully divulged.

A reasonable proxy for the person skilled in the art would be someone filling the same role as the inventor, or a team if there are multiple inventors, but who does not invent or innovate. It is always prudent to ensure the inventors divulge anything in their work that would be considered non-standard and it is better to err on the side of too much instruction than too little.


The EPO provides a fairly concise definition of enablement : Claims are enabled if the disclosure teaches a principle that the person skilled in the art would need to follow in order to achieve each and every embodiment falling within a claim; and the specification discloses at least one application of the principle and provides sufficient information for the person skilled in the art to perform alternative applications of the principle in a way that, while not explicitly disclosed, would nevertheless be obvious to the person skilled in the art.

So, the level of disclosure must be instructional to those skilled in the art, but it is not necessarily required that the document provide some form of proof that the invention works across whole claimed scope.

If claims are broadly drafted and the inventive concept is broad, then the data provided needs to be sufficient to support the proposition that a "principle of general application" exists, and a skilled person would reasonably predict that the claimed invention will work with anything that falls within the broad claim. Usually, in these circumstances, a lot of experimental data will speak for itself but for smaller data sets, it is helpful to provide a reasoned and plausible explanation as to why the principal of general application would be expected to be much broader that that exemplified.

It is not necessary to exemplify or demonstrate every possible alternative within that broad principle of general application, but rather, it is necessary to provide a skilled person with enough information make or do anything that falls within the claim.


Where the invention is narrow, e.g., a single chemical compound or a single method, then the enablement requirement can be satisfied by providing description of how to make that product or how to carry out that method, and a demonstration of the benefit of that compound or method.

As the claims becomes broader, the question of enablement becomes more complex - how closely the experimental detail or data presented enable those embodiments in the claims which are more remote. It is that stage it becomes necessary to consider the burden on a skilled person in putting into practice those embodiments remote from the examples.

Some level of effort is to be expected from the skilled person in working from the examples to achieve the full claim scope, but the burden cannot be an undue one. Clearly, if there is a need to exercise inventive faculty to bridge the gap between the disclosure and the claims, the burden is undue, particularly as the skilled person is deemed to possesses no inventive faculty.

An undue burden would also arise if a skilled person needs to undertake prolonged research, enquiry or experiment to make everything within the scope of the claims. Routine trial and experiment are generally not considered undue burden, but again, that is not always the case since in at least one incidence, the extremely high cost of carrying out otherwise routing testing was considered an undue burden on a skilled person.

As can be seen, the level of data required depends on the nature of the invention, the skill level of the people involved, and the specifics of the work required to arrive at everything within the scope of the claims. The more the applicant can do to reduce that burden, particularly by way of examples or data, the better.


The question sometimes arises - is mere instruction enough to enable an invention? This is usually an issue of support rather than enablement, i.e., questions arise as to the applicant's actual contribution to the art if no experiments have been carried out, however, from an enablement perspective, the use of "prophetic examples" can still be helpful. These detail how an experiment would be carried out, and the expected or modelled results. For enablement purposes, these at least provide a level of instruction which sets out the information that a skilled person would need to attempt the invention, and they can certainly help supplement existing data. Care needs to be taken to ensure that prophetic examples and modelled data are not misrepresented or open to misinterpretation in the specification.


When considering whether the claims are adequately enabled, it can be helpful to consider the sort of claims that might be included in the specification. Some claim types can point to potential enablement issues that should be addressed before filing:

Claim by result - if an invention is primarily characterised in terms of a benefit or a result achieved, then the level of instruction in the description needs to be such that the benefit can be fully realised. It should be expected that any technical features or steps necessary to achieve the benefit will ultimately need to be present in the claims.

Claims with lots of functional language - it is necessary to ensure that every instance of functional language would be as broadly understood by the reader as by the writer. In any case, it is wise to elaborate in the specification as many concrete examples of the functional language as possible.

Parametric claims - claims which include parametric limitations need to be considered carefully. Even well-known parameters should have any variables set out, for instance, if the parameter is temperature or pressure dependent, the temperature and pressure at which the data was acquired needs to be set out (to reduce the burden on the skilled person). Devising new parameters to characterise invention is permitted, but these need to be fully explained to ensure that no undue trial and experiment will be needed by a skilled person to arrive at the claimed invention.

Synergy - A synergistic result is often the winning ticket when responding to inventive step objections, particularly in the chemical arts, but it needs to be borne in mind that the assertion of an unpredictable benefit applies just as well to the applicant's own data as it does to the prior art. If a synergistic result is to be asserted over a broad range of combinations, then good data spread across that range of combinations is essential. Interpolating between synergistic data points is generally fine but extrapolation beyond the examples can be difficult.

Elements of chance - in some cases, inventions involve the element of chance. This is most typical in the biological field for example, where valuable mutated organisms arise from breeding or by irradiation. An element of chance is necessarily prohibitive, provided the need for repetition is not undue. In any event, the enablement of biological samples such as seeds and microorganisms is possible via the Budapest Treaty, which enables deposits of the relevant biological material to be made with the various deposit organisations in order to satisfy the enablement requirements.


Applicants often overlook the addition of comparative data. While this is not relevant to enablement per se, the presence of comparative data is often very persuasive in demonstrating the benefits of the invention.


Applicants should be careful that the data they provide does not contradict the general thrust of the patent claims.

A single data point that contradicts the general principal of operation can lead to the refusal of the patent if left unaddressed. Contradictory data should be checked for correctness - was it the result of an incomplete or blighted experiment, or is it a genuine discrepancy? Bad data should be discarded but where there is a genuine outlier from a general principal of operation, that should be noted and carved out from the claims.


In summary, the amount of experimental detail and data required varies from case to case but provided the key question (whether a skilled person can perform the invention over the whole claim without undue burden and without needing to exercise any inventive faculty) is kept in mind, a reasonable assessment of the adequacy of enablement can be made. There is a degree of subjectivity involved so advice from an experienced patent attorney can be helpful in determining whether more data is needed.

If you found this content from us interesting, we invite you to sign up to our Technology Transfer IP Hub where we regularly release content to assist Technology Transfer and Commercialisation Offices and Research Institutes. This page aims to provide information and resources for the protection of IP developed at academic institutions and publicly funded research organisations.

Sign up to the Technology Transfer IP Hub here.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.