Understanding Eligible vs. Ineligible Software Expenses in SR&ED

One of the biggest challenges for software companies claiming SR&ED is understanding exactly what qualifies. The CRA has strict definitions, and assuming all development work is eligible is a fast track to an audit and a reduced claim.

The Core Concept: Technological Advancement

To qualify for SR&ED, your work must aim to achieve a technological advancement for the purpose of creating new, or improving existing, materials, devices, products, or processes.

In software, this means you must be trying to solve a technological uncertainty.

What is a Technological Uncertainty?

A technological uncertainty exists when it is not known whether a given result or objective can be achieved, or how to achieve it, given the currently available technological knowledge.

If your senior engineers can look at a problem and say, "We know exactly what standard libraries and architectures to use to build this," it is likely not SR&ED.

If they say, "We know what we want the software to do, but we have no idea if the current processing power, algorithms, or database architectures can actually handle it at this scale, and we need to run experiments to find out," then you are likely in SR&ED territory.

Common Ineligible Software Activities

The CRA explicitly excludes certain activities from SR&ED, even if they are difficult or time-consuming:

• Routine upgrades and bug fixes.
• UI/UX design (unless the UI itself involves a deep technological uncertainty, like novel rendering techniques).
• Data collection and routine testing.
• Porting an application to a new platform (unless the platform introduces new, unknown technological constraints).
• Customizing existing software using standard techniques.

Identifying Eligible Activities

Eligible activities typically involve:

• Developing new algorithms to solve complex data processing bottlenecks.
• Creating novel software architectures to handle unprecedented concurrency or scale.
• Overcoming significant integration challenges between disparate, legacy systems where no known solution exists.
• Developing new encryption or security protocols.

The Takeaway

When preparing your claim, always focus the narrative on the technological problem, not the business problem. The CRA does not care that you built a new CRM; they care about the novel caching algorithm you had to invent to make that CRM fast enough.