Policy & Government Relations SR&ED Basics

SR&ED in Software Development: What is System Uncertainty?

July 10, 2023 Janette Conway

* This information is presented for educational purposes only and does not constitute legal advice. You should retain legal counsel if you require legal advice regarding your individual tax situation. *

code on a computer screen — SR&ED in Software Development: What is System Uncertainty? (Photo Credit: Markus Spiske via Pexels.com)

The software development industry has been advancing rapidly in recent years, and we rely increasingly on computers in our day-to-day lives. These advancements can often result from Scientific Research and Experimental Development (SR&ED). Therefore, it is increasingly common for software companies to claim the SR&ED investment tax credits (ITCs) for the work they complete, which advances the knowledge base in their fields. This article will discuss SR&ED eligibility within the software industry and how system uncertainty may play a big role in your company’s SR&ED.

Software Development

Within the software industry, it can be difficult to separate the business/commercial goals of the company from the SR&ED goals. More specifically, we have observed confusion when differentiating between technological uncertainty, which must exist for SR&ED to occur, and general complexity, which often occurs in software development, especially during software integration projects.

System Integration: the process of linking together different computing systems and software applications physically or functionally, to act as a coordinated whole.¹

Technological Uncertainty

To effectively discuss System Uncertainty, we must first remind ourselves what the Canada Revenue Agency considers a Scientific or Technological Uncertainty.

The CRA defines technological uncertainty in the SR&ED Glossary:

Scientific or technological uncertainty means whether a given result or objective can be achieved, or how to achieve it, is unknown or uncertain due to an insufficiency of scientific or technological knowledge.²

In our article The 5 Questions For SR&ED – Intent and Evolutions we discuss how the term uncertainty and others have evolved over the years since the SR&ED program was initiated. We will now look at this definition in relation to two key CRA eligibility documents: The Eligibility of Work for SR&ED Investment Tax Credits Policy (2015) and the Guidelines on the Eligibility of work for scientific research and experimental development (SR&ED) tax Incentives (2021).

System Uncertainty

If you are unsure, or doubt, that the computing systems and/or software applications will work together, then it is not inherently true that a technological uncertainty exists. Uncertainty is more than complexity and must be established by demonstrating the existing knowledge base is insufficient; however, the Policy also describes a form of technological uncertainty called system uncertainty:

System uncertainty can arise from or during the integration of technologies, the components of which are generally well-known. This is due to unpredictable interactions between the individual components or sub-systems. It may be difficult or impossible to predict how the integrated system will perform due to unforeseeable adverse interactions. The uncertainty here is not in the individual modules or components but in the modules or components acting as an integrated system. The attempt to resolve these uncertainties by a systematic investigation or search can lead to technological advancement.³

The Canada Revenue Agency (CRA) is aware that software development often results in unanticipated issues during the integration of different technologies. Believing that software integration is going to be difficult or may not work will not meet the definition of a “technological uncertainty”; however, if during the software integration, you encounter unpredicted and adverse interactions which impact how the technologies act together as an integrated system and you can demonstrate how this meets the definition of technological uncertainty (incl. knowledge base) then those interactions could be considered system uncertainties, and be eligible for SR&ED.

SR&ED Eligibility Policy (2015)

The “Eligibility of Work for SR&ED Investment Tax Credits Policy” is an important source of information for SR&ED claimants, and even though the policy was archived in 2021, it is still being used in the Tax Court of Canada. One of the most recent examples is the case of Dave’s Diesel Inc. v. The Queen (2022), in which the judge specifically used the 5 questions of eligibility to determine if the Appellant’s work was SR&ED eligible.⁴ The 5 questions were initially put forth by Justice Bowman in 1998 and were then included in the 2015 SR&ED eligibility policy as the main method to determine if work carried out was eligible for SR&ED. The very 1st of these 5 questions asks “Was there a scientific or a technological uncertainty?”⁵

In the Eligibility of Work Policy (2015), there is a detailed explanation of the difference between technological uncertainty and complexity, stating that:

Doubt about the business or commercial success of the material, device, product, or process being developed is not a scientific or technological uncertainty.

Furthermore, complexity does not necessarily mean the existence of technological uncertainty. The size and complexity of a project by itself does not justify that the work performed in that project falls within the definition of SR&ED. Likewise, the fact that a large and complex system was developed cannot support the inference that an uncertainty existed.⁶

SR&ED Eligibility Guidelines (2021)

The Eligibility of Work for SR&ED Investment Tax Credits Policy (2015) has since been archived and was replaced by the Guidelines on the Eligibility of work for scientific research and experimental development (SR&ED) tax incentives (the “Guidelines”) which do not specifically define system uncertainty. However, we can still conclude from the text that this form of technological uncertainty remains SR&ED eligible.

Similar to the glossary, the basic requirement for work to be considered SR&ED-eligible remains the same, which is the identification of technological uncertainty and experimental development work geared towards a technological advancement:

The recognition that scientific or technological uncertainty exists marks the starting point for the SR&ED work, while the advancement is the targeted outcome of the work. Therefore, an attempt to resolve scientific or technological uncertainty is an attempt to achieve scientific or technological advancement.⁷

In the context of experimental development, the work is conducted to achieve technological advancement. In other words, “the work is conducted for the generation or discovery of knowledge that advances the understanding of technology.”⁸

Note that the wording in the current Guidelines state that “the recognition of a […] technological uncertainty marks the starting point for the SR&ED work” – this makes clear the gap in your knowledge base must be clearly identified to delineate between development and experimental development work. The guidelines then state the following:

When developing new or improved materials, devices, products or processes, problems can occur when there is a need to achieve a set of features, functions or capabilities, while being restricted by a set of constraints. When the technological knowledge available to your business is insufficient, attempting to resolve these problems through experiment or analysis will result in the generation of new technological knowledge.⁹

If a problem is encountered which cannot be solved with the information from the existing knowledge base available to the company then an advancement (to your knowledge base) is required. In the SR&ED world, two groups of uncertainties may be encountered: those that are known (official starting point) and those that are expected but unknown (system uncertainty). Technological uncertainties that arise during the integration of software and technologies are often discovered after the process has already begun. Be careful to identify the point at which you realize that insufficient knowledge is available to your organization.

To assist taxpayers, the guidelines list the possible characteristics of problems that may suggest the technological knowledge is insufficient:

Existing design methods are not applicable;

Requirements or specifications do not conform to existing standards;

Too many variables or unknowns;

Parameters or conditions are outside of the normal operating range;

Nature of the problem is evolving;

Data is not readily available;

There are interlocking constraints.¹⁰

The more of the above characteristics that are true for a problem, the easier it will be to prove that the problem is/was a technological uncertainty.

Tax Court of Canada

There are a few interesting Tax Court of Canada rulings that specifically discuss the concept of system uncertainty. We discussed the case of Dave’s Diesel Inc. v. The Queen (2022) above, but we have delved further into the case of Buhler Versatile Inc. v. The King (2023) and Clevor Technologies Inc v. The Queen (2019) as they are specific to the software development industry.

System Uncertainty in Buhler Versatile Inc. v. The King (2023)

Buhler Versatile Inc. v. The King (2023)
- SR&ED Education Resources Legal Ruling Summary – Member access only

In the case of Buhler Versatile Inc. v. The King (2023), the Appellant, Buhler Versatile Inc., sought to appeal the federal Income Tax Act determination made by the Minister of National Revenue pertaining to the Appellant’s 2005 taxation year. The Minister disallowed the Appellant’s 2005 scientific research and experimental development (SR&ED) claim for qualified expenditures totalling $3,591,220 with respect to seven projects. The Appellant is an agricultural equipment manufacturer located in Winnipeg that specializes in the manufacture of agricultural tractors. The bulk of the SR&ED expenditures incurred in the Appellants 2005 fiscal year were concerning their 5th project, in which the Appellant aimed to create a line of high-horsepower 4WD tractors which met Tier II emission standards and were suitable for agricultural and commercial construction (e.g. scraping/earth-moving/levelling) applications.

The 5 questions initially put forth by Justice Bowman in 1998 and written into the 2015 Eligibility of Work for SR&ED Investment Tax Credits Policy, were used in this case to determine the Appellant’s eligibility. To this day these questions are the main method used to determine if the work carried out is eligible for SR&ED. As the Eligibility of Work for SR&ED Investment Tax Credits Policy specifies, “it is important to distinguish between a systematic approach to carrying out work and the approach that is a systematic investigation or search called for in the definition of SR&ED.”¹¹

In examining whether the Appellants work contained technological uncertainty(ies), Justice Bowman’s first question, the judge determined that the technological uncertainty, in this case, fit under the description of a system uncertainty as all of the parts needed to function successfully together to achieve the Appellant’s objective:

[57] I am of the view that the technological uncertainty, in this case, fits squarely under the description of a system uncertainty. In other words, the integration of non-trivial combinations of established (well-known) technologies and principles carried a major element of technological uncertainty. When all the individual parts were combined, their individual uncertainties were merged into a system uncertainty and the system uncertainty was the entire tractor. All of the constituent parts needed to function in unison to achieve the appellant’s objective.¹²

The judge also noted that:

[60]… While the techniques and testing methods were known, the resolution of this system uncertainty was not reasonably predictable, as can be seen by the number of directions the appellant went in its efforts to resolve the cooling and coupler challenges. The system uncertainty was so unpredictable that the appellant had to build a customized torsional coupler in order to continue with its larger goal of incorporating the QSX-15 engine in the process of building a Tier-II-compliant 535-HP tractor.¹³

The Appellants project aimed to take existing technologies and integrate them. The integration of these technologies was likely to result in unexpected adverse interactions that the Appellant could not “reasonably predict”. The judge ruled that the Appellant’s “activities with respect to the 535-HP tractor in project 5 (4WD Phase D Tier II High HP) constituted SR&ED in the 2005 taxation year”, and they were therefore successful in proving to the court that their work was SR&ED eligible.

System Uncertainty in Clevor Technologies Inc v. The Queen (2019)

Clevor Technologies Inc v. The Queen (2019)
- SR&ED Education Resources Legal Ruling Summary – Member access only

In the case of Clevor Technologies Inc v. The Queen (2019) the Appellant, Clevor Technologies Inc., appealed a reassessment of their SR&ED investment tax credit claim for the 2013 taxation year. The Appellant claimed two projects were eligible for SR&ED investment tax credits during this taxation year, the first of which involved the integration of the sophisticated project management software application the Appellant developed prior to 2013, the Clevor Schedule Optimizer (CSO), with a new third-party software that provided “front end” to the customer in the linked operation of the two applications, Primavera 7. The program interface for the CSO was changed which, in turn, forced the Appellant to change the code of the CSO for the two programs to integrate:

[7] However, in or about early 2013 Oracle updated its “application programming interface” (API) code for its new version of Primavera being Primavera P7. This change blocked CSO from integrating with Primavera P7, pending adaptations of CSO’s code. While Oracle had published, for reference by software companies with products integrating with Primavera, an explanation of its API changes, that published explanation was insufficiently comprehensive to permit the Appellant to readily ascertain required code changes for CSO.¹⁴

The Appellants project aimed to take the existing technologies, CSO & Primavera P7, and integrate them. The integration of these two technologies was likely to result in unexpected adverse interactions as the Appellant could not “readily ascertain required code changes”. It may have been difficult or impossible to predict how the integrated systems would perform as an integrated system. The Appellants’ project hypothesis was as follows:

[t]he hypothesis generated was that the changes made to the API that affect [the Appellant’s] integration could be determined if developers systematically tried various combinations of XML items [an aspect of API code] and added/removed different item fields to eliminate the errors, and warnings, generated when a partial XML file was used to update a project in Primavera 6. The knowledge gained from this systematic investigation improves our understanding of the new schema file and help[s] [the Appellant’s] future integration work.¹⁵

The 2015 Eligibility Policy states that a systematic investigation must include the following aspects:

formulating one or more hypotheses designed to reduce or eliminate the uncertainties;

planning and executing the testing of the hypotheses by experiment or analysis; and

developing logical conclusions based on the results or findings of the experiment or analysis.¹⁷

Per the Guidelines, a systematic investigation must include the following aspects:

Generating an idea consistent with known facts, which serves as a starting point for further investigation toward achieving your objective or resolving your problem. Your idea may be expressed as a possible solution to a problem, a proposed method or an approach. This can be referred to as a hypothesis.

Testing of the idea or hypothesis by means of experiment or analysis. The idea can evolve and change as a result of testing.

Developing logical conclusions based on the results or findings of the experiment or analysis.

Keeping evidence that is generated as the work progresses.¹⁸

The only element added to the current Guidelines to determine whether a systematic investigation occurred is whether or not documentation was kept along to way. The main element in determining the process has remained the same.

It is entirely possible and likely that the Appellant encountered system uncertainties during the integration of CSO and Primavera 7; however, the Appellant was not successful in this case as the court ruled they did not approach this work in a manner consistent with a systematic investigation, nor did they document it adequately enough to prove their work was SR&ED eligible. See our series of articles on Trial and Error so you don’t make these same mistakes with your projects: Trial And Error: Separating SR&ED From Blind Approaches (Pt. 1 of 2) & Trial And Error: Separating SR&ED From Blind Approaches (Pt. 2 of 2).

More Examples of Legal Rulings Relating to System Uncertainty

Note: Only SR&ED Education & Resources Members are able to access our Legal Ruling Summaries. For more information on membership and the Comprehensive Guide to SR&ED please reach out to us.

Highweb & Page Group Inc. v. The Queen (2015)
- SR&ED Education Resources Legal Ruling Summary – Member access only

In this case, the Appellant aimed to enhance their company’s web content management system by “universalizing” its compatibility with other commercial operating system platform programs and commercial web-based server platforms. The Appellant maintained that “system uncertainty” existed, however, they were not able to prove that a systematic investigation took place, nor that any technological advancements were made, and therefore the judge dismissed their appeal.

Gabie v. The Queen (1998)
- SR&ED Education Resources Legal Ruling Summary – Member access only

In this case, the Appellant aimed to prove that software engineering was a valid form of engineering, and that he was eligible for the related Overseas Employment Tax Credit (“OETC”) he had applied for. At the time of this ruling, 1998, the OETC was a tax credit program available to individuals who were resident in Canada but carried on business outside Canada on a resource, construction, installation, agricultural, engineering or prescribed activity.¹⁹ The Appellants claim for OETC was denied as, at the time, software engineering was not widely accepted as a form of engineering or science. The Appellant used the definition of Scientific or technological uncertainty within the SR&ED program at the time to defend his position, emphasizing that the technological uncertainty experienced within software development, specifically called “system uncertainty”, “exists only if the integration is not achieved through routine engineering and requires changes to the basic design of the underlying technologies…”. While this case was not related to the SR&ED ITC program, it does show that even before the pivotal case of Northwest Hydraulic Consultants v. The Queen (1998), in which Justice Bowman posed the 5 questions to SR&ED eligibility, system uncertainty has been an accepted form of technological uncertainty.

Conclusion

Although the CRA’s eligibility policy has been archived and the new SR&ED eligibility guidelines do not explicitly reference system uncertainty in SR&ED, system uncertainty is still considered SR&ED eligible.

While we do not recommend relying on system uncertainty alone in an SR&ED technical narrative, the possibility of system uncertainties occurring should not be overlooked by those working in the software industry. It is always a good idea to be prepared in case your research goal pivots in a different direction. Thus, keeping track of the work completed and all the uncertainties encountered is essential. Contemporaneous documentation is the best tool you can use to substantiate your claim in the case of a review. Our previous article, Why do I need contemporaneous documentation for SR&ED? delves further into this topic.