Comparing the Canadian and European Approaches to the Patentability of Computer-Implemented Inventions

June 16, 2021

By Paul Horbal, Madelaine Lynch, Denis Keseris and Christopher Kowal

In March of this year, the Enlarged Board of Appeal reaffirmed the European Patent Office (EPO)’s approach to the patentability of computer-implemented simulations.1 While the EPO continues to follow a “technological contribution” approach to patentability, Canada has recently rejected a similar “problem-solution” approach in favour of an approach based on purposive construction of the claims. This article looks at both approaches and how they apply to computer-implemented inventions.

European Patent Office Approach

The EPO has a two-hurdle approach to patentability. Under Article 52(1) of the European Patent Convention, “patents shall be granted for any inventions, in all fields of technology, provided that they are new, involve an inventive step and are susceptible of industrial application.” The first hurdle is patent eligibility, which is assessed without reference to the prior art. Article 52(2) EPC lists items that are not considered eligible inventions which includes, amongst other abstract ideas, programs for computers. However, for computer-implemented inventions this first hurdle has become relatively easy to clear, and can be satisfied by including the use of a computer in the claimed subject-matter. This was deemed the “any hardware” approach (T 0253/03) as a single technological feature is sufficient to satisfy the first hurdle.

It is only at the second hurdle, where the so-called COMVIK approach (from the EPO Board of Appeal decision T 0641/00) becomes relevant. The COMVIK (two identities) approach focuses on the “inventive step” requirement. Here, the prior art is compared to the claimed subject-matter in order to distinguish the novel features from that of the closest prior art. The caveat is that only those features that contribute to a technical solution (i.e., a “further technical effect”) of the objective technical problem (i.e., the problem solved as compared to the closest prior art) are considered. Any non-technical feature of the invention is effectively considered as prior art for the purposes of assessing inventive step. Most of the action therefore happens in determining whether a feature contributes to producing a further technical effect.

Legislators and courts are reluctant to define the phrases “technical” or “technology” as technology is an ever-evolving field. Setting a definition for what qualifies as “technological” today may inadvertently preclude the technological advancements of tomorrow. Yet, it is this moving target that makes interpreting the technological features of the claimed subject-matter so challenging.

The Enlarged Board of Appeal (EBA) stated that a “technical system or process implies that an object is created or a process is run with some purpose based on human creativity.”2 Put another way, the EPO draws a distinction between predicting the state of a system, and controlling something by way of the prediction. Consider weather, for example, which was deemed to be a physical system that can be modelled to show how it works, but not a technical system that would be improved by the skilled person. Indeed, even a system that improved the reliability and predictability of weather forecast data was held not to be a technical problem. However, if that same weather forecasting data is used to automatically open or close window shutters it may be considered to contribute to the technical character.3

In the context of simulations, the EBA determined the COMVIK approach is to be applied as with any other computer-implemented invention. Importantly, it was noted that what the simulation is modeling is not important and may be technical or non-technical. Rather, the effect of the simulation (the solution to a technical problem) is the relevant feature and must go beyond the simulation’s straightforward or unspecified implementation on a standard computer system.4 With computer-implemented inventions becoming increasingly prominent, the inventive step may become more challenging to overcome. The use of a generic computer technique applied to a non-technical process may not satisfy the inventive step as the generic computer technique does not rise above the prior art.  

Canadian Approach

Following the Federal Court decision in Yves Choueifaty v Attorney General of Canada,5 the Canadian Intellectual Property Office (CIPO) released a guidance document (PN2020-04), which supported the use of a purposive construction approach in construing the essential elements of a claim. Determining the essential and inessential elements of a claim using the purposive construction approach presumes that all elements of a claim are essential unless there is evidence to the contrary. By doing so, the “technological solution to a technological problem” and the “problem-solution” approach are no longer applicable standards in Canada.

Similar to the EPO, in Canada an invention must be new, non-obvious, useful and must fall within the definition of invention. The invention requirement is similar to the first “hurdle” of patentability criteria described above. However, prior to Choueifaty, computer claim elements were routinely construed as non-essential, leaving only a mere scientific principle or abstract theorem, and thus not patentable subject-matter (see our article here).

While a mere scientific principle or abstract theorem remains unpatentable subject-matter in Canada, they can be construed as essential elements together with elements that have a physical existence. However, unlike at the EPO, the mere fact that a computer is included in the claim is not a guarantee that the claim will be considered directed toward patentable subject-matter. If the computer is merely executing an inventive algorithm, but the computer and algorithm cannot be said to cooperate together to form a single invention, the claim would not be considered patentable subject-matter unless some other element of the claim causes “a discernible physical effect or change.” However, if the algorithm improves the functioning of a computer, the computer and algorithm can be considered to cooperate to form a single invention and thus be patentable subject-matter.

It is at this first stage of the analysis that Canada differs from the EPO. In Canada determining patent eligibility requires the claimed subject-matter to be construed, and consideration as to whether the essential elements of the claim as construed “cooperate” to form a single invention, or else manifest some physical effect. In contrast, the EPO’s “any hardware” approach sets a low bar for patent eligibility and does not consider the essential elements until the inventive step (obviousness) analysis.

The claim construction at the patent eligibility stage is carried throughout the analysis of novelty, non-obviousness and utility. In Canada, the essential and non-essential elements do not align with the technical and non-technical features of the EPO approach. Instead, consideration is given to the intent of the inventor; absent evidence to the contrary, all features of a claim can be considered essential and thus relevant to the obviousness assessment.

Although, the claims are not construed to look only at the technical features, they may nonetheless produce a similar outcome in that an invention which has no physical existence or physical effect or change would not be considered patentable subject-matter.

Indeed, in the first example provided by CIPO:

A computer-implemented method of analysing data from seismic measurements comprising:

  • Performing seismic measurements;
  • Receiving the data from the seismic measurements;
  • Processing the data on a computer using algorithm X; and
  • Displaying the results of the analysis

All elements of the claim are construed to be essential. Under the COMVIK approach the computer-implemented method would satisfy the technical requirement, but so too would the seismic measurements as they are based on an interaction with physical reality. Thus, in Canada and the EPO this would be considered patentable subject-matter. The ultimate decision on patentability would thus depend on the comparison to the prior art.


Although the outcome may be similar much of the time, the Canadian and EPO approach to computer-implemented inventions are quite different. At the EPO, the heavy lifting is done in the “inventive step” stage. In Canada, much of the analysis centres on the assessment of whether elements cooperate to form a single invention. While the methods may differ, ultimately, both offices require the computer-implemented inventions to go beyond the mere use of a generic computer.

1 EPO Enlarged Board of Appeal decision no. G 0001/19 (

2 G 0001/19 at para 47.

3 G 0001/19 at para 131.

4 G 0001/19 at para 51.

5 Yves Choueifaty v Attorney General of Canada, 2020 FC 837.

Subscribe to our newsletter

You can unsubscribe at any time. This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

This site is registered on as a development site.