In the Scientific Research & Experimental Development (SR&ED) tax incentive program, the pursuit of a technological advancement (TA) is about the attempt to gain some small, incremental improvement in the face of technological uncertainty, outside the boundaries of known or standard approaches, and through experimental means. It’s about taking baby steps, and seeking small gains – just like a child learning to walk. And that, too, is an experimental process.
The official definition of experimental development is “work undertaken for the purpose of achieving technological advancement for the purpose of creating new, or improving existing, materials, devices, products or processes, including incremental improvements thereto.” (There’s more, and it bears careful studying, but then, most of the literature around the SR&ED program is fundamentally about either qualifying or “scoping” – what you can claim, and how much.)
There was a time when the pursuit of TA was the first criterion that was mentioned in any discussion about SR&ED. (Remember that any discussion of TA implies, also, the pursuit of the advancement, whether or not it is achieved.) Certainly, it is a key criterion. In recent years, I’ve made it a point to mention technological uncertainty first, because the identification of TU is the first point at which SR&ED becomes possible, and the end of TU signals the end of SR&ED. You could think about the TU criterion as a pair of brackets, the beginning and the end, within which you have to be pursuing a technological advancement by experimental or scientific means.
However, Technological Advancement (TA) is a problematic criterion for many people, because it is so often confused with business innovation, and with features and functionality. Adding what used to be called “bells and whistles” to a product, for example, will not qualify as the pursuit of TA, unless you are attempting to make changes to the underlying technology that affect how it works, not just what it does.
There is always, with SR&ED, an expectation that a claimant is pursuing the development of a claim within the spirit of the law, and not attempting to “game” the system or pull a fast one. While such behaviour undoubtedly occurs, I’d like to believe that more common errors arise out of various misunderstandings relating to this criterion.
It is not helpful, for example, to engage in hype-speak or marketing prose on the subject of the technology advancement. Claiming to have achieved “the most advanced product in the marketplace” demonstrates nothing about the process of pursuing SR&ED – and it both annoys the reviewers and provokes them, because it demonstrates a fundamental misapprehension of the SR&ED criteria.
If the capabilities achieved or sought through the process of SR&ED represent an advancement in the underlying technology, then it should be possible to describe those advancements in scientific or technical terms, and preferably, quantifiable terms. Marketing hype just drops a really fat and obvious fly into the soup, and it makes the whole dish unpalatable.
One of the clues to the pursuit of TA is that it often results in transferable and reusable knowledge. It is not uncommon in this context that a highly trained person may actually discount or dismiss the SR&ED in their project because they feel that the advances sought don’t represent an “absolute advancement” in the world-wide knowledge base relevant to the technologies and techniques involved. That’s desirable, but not essential.
In fact, it is true that the business context of the company and its internal know-how or “technology base” also need to be considered. The company may yet have learned something from doing the work. Where there is a real gain in the technical knowledge of the company – its technology base – and if technological uncertainties have been addressed or reduced in doing so, then the work was SR&ED.
There is no requirement that advancements claimed have to represent a giant technological leap – just some small gain, a lesson learned, a problem better understood, or even a failure that can be avoided in the future. The pursuit of advancement is generally the pursuit of knowledge or know-how, and even wisdom of sorts, one baby step at a time.
Bruce Madole
Sometimes a claim is just too small, too weak, or too poorly supported to be worth the effort and cost of pursuing it. Occasionally, however, a small claim represents an opportunity that it could cost you dearly to overlook. When is a small claim not too small? When it’s a seed project.
Every claim has a cost associated with preparing it. All of the staff engaged in the SR&ED process cost money– the in-house technical and financial people who have to be interviewed, who spend time looking for eligible work or in reviewing their own notes, records and files to find supporting evidence for work that is being claimed. Even where the claim preparation is being assisted by consultants on a purely contingency basis, there is still an up-front additional cost (i.e. the value of the time of internal staff) when meetings and interviews have to be arranged in defining and scoping claims.
On any claim, there comes a point when certain opportunities are weighed against the effort and cost to prepare them, and the key words are then uttered: “too small to bother with” or “not worth it”. The decision is usually made based on the value of the claimable costs within the given fiscal period, relative to the effort and cost in preparing a defensible claim document for it, and sometimes, the amount of time you have left before filing deadlines. Before you make a final decision about such projects, you would be well-advised to take another look, and decide whether one of these small projects could represent the “seed” of something much larger and more important.
I call them seed projects because they represent, usually, the earliest opportunity to recognize key technological uncertainties (TU), and because the early identification of such uncertainties paves the way for SR&ED work to occur in seeking to resolve them. This is an important element in maximizing your ability to claim.
How do you recognize a seed project? What do you look for? How can you recognize a seed project for the massive tree it may become, instead of the tiny seed of effort and cost it is now? Here are three early indicators of a seed project:
* Project is identified by your core technical primes: When the people you pay to be your best and brightest minds are poking away at a problem, even if it’s in something of a “skunkworks” mode, you should pay attention to that. The very definition of TU is that it occurs when the trained professionals cannot succeed with standard approaches.
* The key TU relates to an emerging, expanding or desired line of business: When the barrier or obstacle that is studied is related to something that the company really wants to pursue or is moving toward, then there is a prospect that the uncertainties and desired advancements will assume much greater importance, and may eventually even become mission-critical. The potential for downstream investment is much greater if the TU relates to a corporate priority.
* Early efforts or earliest explorations are technically failures, or unpromising: If people are working on a problem, and the approaches have failed to some degree, then persistence is its own clue. You need to ask, what’s at the end of this rainbow?
These factors don’t mean that you should always claim such small projects – only that you could lose (or diminish) tremendous opportunities in failing to do so. Early identification is about maximizing your potential claims. It’s a good thing to take a longer view of the process, and to avoid becoming a prisoner of one’s own cost-benefit analysis.
Bruce Madole
Of course, no discussion of TA would be complete without talking about “novelty”. Novelty is what the Canada Revenue Agency (CRA) refers to when we make an argument that a certain chunk of work must be a technological advancement because that work entailed doing something that has never been done or attempted before.
The first response to such claims is generally … so what? It is possible to seek or achieve a commercially unique result using purely known technologies and standard approaches, and without addressing a single technological uncertainty. The CRA’s position on the novelty argument is that originality is not necessarily the same thing as technical innovation.
Still, just because a thing is novel, that doesn’t make it ineligible either – it’s just that the argument, by itself, doesn’t gain you much traction for your claim. Further clouding the issue, it is common for the novelty argument to be linked, at the same time, with advancements that are based on commercial progress, with features and functionality presented in a commercial sense rather than at the level of the underlying technology. The two issues often pop up together and it becomes necessary to sort out two tangled misapprehensions instead of just one. Having first restored a focus on the technology discussion, it then becomes possible to explore the novelty issue from a technology perspective. Exploration is definitely necessary.
Although novelty may not be the definitive argument for technological advancement, it does offer intriguing possibilities, and may well be an indicator of possible SR&ED. I suspect that most science reviewers confronted with this argument would probably wish to say something like, “Yes, yes – it’s neat what you did. Now can you tell me why and how you did it, and what problems you encountered, at a technology level?”
If a thing has never been done before, the question you need to ask is, “why not”? Is there some technological uncertainty (TU) about the attempt? Has nobody ever tried to do it? Or have lots of people tried and failed in various ways? (If lots of people have tried and failed, there is presumably good evidence for the existence of technological uncertainty.) Moreover, a lack of generally available knowledge may well form a part of the TUs confronted in such a project.
Proprietary knowledge is not “generally available”, and is not commonly disclosed to the technical staff of a company that is trying to use a given technology. Where the novelty argument truly exists, at a technology level, it would not prove startling or uncommon for the required technology information to be either unavailable, or not widely known. To make such a claim, you need to explain where and in what respects the needed knowledge was lacking, and what you had to do, experimentally, to overcome the problem. You still have to talk about technological uncertainties and about experimental development work that was done to address them and to pursue the advancement.
It’s a matter of moving from “so what?” to “and then what?” and of understanding that novelty is not so much proof of an advancement as it is another indicator of technological uncertainty.
Where the new or novel thing actually does represent a technological advance, because you have addressed technological uncertainties through experimental means, then there should be eligible work.
Bruce Madole
Project complexity does not equate to technological uncertainty, nor does overcoming complexity translate to technological advancement. I’ve learned this from hard experience, and I will share why.
From time to time, a company will find themselves wishing to claim work from a project that was very large and extremely complex. Often, such projects may involve systems integration and/or engineering activities embodying multiple technologies and even multiple locations, with stringent technical requirements. Standards or regulatory compliance may also be a factor. (Regardless of the size of the company or its project, you may recall projects of your own that involved highly complex problems. Complexity is not exclusively a problem for large companies.)
It is not uncommon for the claimant to point to all of the contributing factors in a complex situation and to argue that the complexity is, in itself, evidence of technological uncertainty. That’s not really true, though. Complexity is evidence of business risk, and extreme complexity may contribute to extreme risk, but a reviewer will remind the claimant that engineers and systems developers are paid well to cope with complex challenges – it’s what they are paid to do, and is a part of the job. The reviewer’s second rebuttal flows from this observation, the argument that dealing with complexity is “business as usual” or “routine engineering”. And it is certainly true, though not necessary, that many complex problems may be overcome by employing purely routine approaches.
Once the “R” word has been uttered, you are on a slippery slope. It can prove really, really difficult to overcome a reviewer’s initial perception that the work performed was routine in nature. If you are truly lucky, the reviewer may ask the question that should have been asked beforehand, and answered in the claim itself, which is: “how did the complexity result in, or contribute to, technological uncertainty?” Were there areas, amid all of the complexity, where the problems encountered went beyond the standard approaches or generally available knowledge? Were there situations where you tried to overcome such problems systematically and experimentally?
Complexity ought to be seen as an invitation to dig deeper into the work. Like novelty, complexity is an indicator of potential SR&ED, not a proof of its existence. It provides a context in which systems uncertainty, for example, may become possible. However, to prove that SR&ED work has actually occurred, you have to go beyond the complexity itself to identify and document the areas of technological uncertainty, and the experimental approaches that were employed to overcome those challenges.
Bruce Madole
Perhaps you are familiar with the phrase “the sound of one hand clapping”. It’s all very mysterious and evocative. I’m not particularly mystical or meditative, but I’m going to use this phrase in a discussion of two opposite extremes in the preparation of an SR&ED claim. My thesis is simple -- a SR&ED claim needs to strike a balance between a description of the qualifying technical work, and the costs that may be reasonably associated with that work. Putting too much emphasis on either the financial or the technical aspects of the claim will endanger the claim.
Let’s suppose, for the sake of this analogy, that the financial and technical aspects of the claim represent one hand each, whether clapping or otherwise. Put yourself in the position of the Science Reviewer who has been asked to review a claim. In our first extreme example, the claimant has assembled a claim that provides exhaustive coverage of all of the costs associated with doing the claimed technical work, but those costs are not linked to any description of the technical work performed. (Of course, this is impossible, you might observe, because even in the new short form you have to fill in something about the technological uncertainties, technological advancements, and work done, or you won’t have a claim. It’s an extreme example, as I said, but please also reflect that if you choose to write about the wrong things, such as business risk or the routine development of commercial functionality, then the reviewer will find his or herself just as challenged for information as though you had written nothing at all.) In this example, there’s no way to tell which costs are related to the SR&ED, or whether any SR&ED has in fact occurred, because the claim lacks the fundamental details that would transform the list of costs into a real claim, by either qualifying the eligible or eliminating the ineligible.
The costs in this first example would be classified as “unsupported”. That’s not “one hand clapping” – it’s the sound of one hand waving goodbye to a SR&ED claim, simply because there was no effort to deal with the technological basis for claiming.
In the second extreme example, the project description is richly endowed with details of technological uncertainty, the pursuit of technological advancement, and a detailed history of the work performed. But there are no costs associated with doing that work. No time spent. No equipment purchase or materials consumed or transformed. The work might as well have been done by ghostly volunteers.
That sounds absurd, and of course it is. If you had all of that information available about who was doing what, and when and why they did it, there would have to have been some kind of record, right? That would have been evidence that the work was going on? There’s no point in a claim if you can’t link to any costs associated with doing the work: in the absence of costs, you are left with some pretty impressive volunteering. The one hand, in this case, is not clapping – it’s the sound of a reviewer scratching his head in puzzlement, wondering why you bothered to prepare anything at all, if you didn’t want the money.
Claiming SR&ED work is about claiming costs that are “commensurate with” the work performed, and vice-versa. It means, in practical terms, that you have avoided both of the extremes we have been writing about, and found a reasonable balance between what is claimed, and what isn’t, and that those reasonable eligible costs are quite reasonably linked to the eligible work.
In the end, claiming SR&ED is all about getting the balance right, and linking the effort to the eligible costs. Get that part right, and you will hear something not all that Zen-like or mysterious: the sound of two hands clapping. It’s called applause.
As I discussed in the previous post, these processes need to be extremely lightweight, because most companies are in the “business” business, which is to say, manufacturing, or computer programming, or tool and die work, etc., or whatever business they happen to be in. SR&ED is supposed to help by mitigating the costs of technical risk, by supporting the development of new and improved products and services, and by providing a financial incentive for the pursuit of new knowledge, experimentally.
The initial efforts to develop a SR&ED Framework, and the necessary supporting processes, may mean that the SR&ED program strives first for effectiveness and then secondarily, for efficiency. SR&ED process thinking entails a paring down of anything that isn’t the minimum required effort to achieve that desired effectiveness, because your effectiveness depends on the full cooperation of others who do not, in all likelihood, share your concerns. They have other jobs to do – technical, financial, etc – and what you are asking of them will require cooperation and effective “buy in” to justify what will be, at the very least, an unusual demand on their time. Keeping things lightweight, and trimming down the level of interruption or extra effort, where you can, is critical for the success of your framework and processes.
Ask yourself, what would be the easiest way to do this?
Well, the easiest way to do SR&ED is where the technical experts in a company understand what is eligible, and they identify that eligible work instantly, as it arises. Having identified eligible SR&ED work, the technical experts do the necessary technical work and they keep sufficient documentation (of both the work and the costs) so that the work can be claimed with minimal effort. From a program level, that means that the ownership and direction of the science part should be led by the technical experts in a company. The financial and operational parts – such as scope of effort, costs, contractual eligibility, and claim preparation – may be a combined effort of a project management office and SR&ED Program Management office, while the tax filing will be led by Finance and Taxation teams. Moreover, these processes, which sound as though they are designed for big corporations (true) are also scalable down to small companies. If you think of the SR&ED process as a relay race, the difference between the large and small process is the number of “handoffs” between individual roles, and the effort to be sure that nobody “drops the baton” while you’re trying to make the finish line.
If there was to be a fourth process required, it should be the process for ensuring that you have understood and extracted the necessary learning – about the technical work, and about the SR&ED process itself. This would be the process for circling back, and looking at lessons learned, and the part of the process where you start re-examining the Framework (your spider web) to see which bits you can improve, going forward.
That would be, thinking about SR&ED and …
This article talks about SR&ED process thinking, first by examining the concept and characteristics of a framework for pursuing SR&ED, and then, secondly, by reflecting on the processes that might be needed as a part of such a framework.
I think a spider web is a useful paradigm for building SR&ED-supportive processes within your company. Have you ever watched a spider spinning a web? One tiny thread patiently joined from one point to another, until eventually … coverage.
I don’t want to extend this example tediously, but there are three aspects to the spider’s web that are worth commenting on: linkages, framework, and coverage.
First, you start with linkages – one thing links to another, and to another. Technical SR&ED work needs to be linked to the evidence of the work, which needs to be linked to the costs of doing that work. Creating those linkages, the chain of evidence from the SR&ED and supporting work to the costs, is something that you should think about frequently. As I see it, this “thinking about SR&ED” is a matter of habit. You get used to thinking about “SR&ED and …something” -- SR&ED and project management, SR&ED and project costing, SR&ED and payroll , SR&ED and … nearly everything.
From the gradual accumulation of linkages, one develops a framework – a structure that needs to be both lightweight and comprehensive. Both of these are essential attributes. The framework needs to be lightweight, because SR&ED is about supporting your business, and it is the business that drives the technical work. The SR&ED framework also needs to be comprehensive, because it is meant to help you maximize your SR&ED claims, so as to take best advantage of the benefits for which you may be entitled.
Finally, your framework needs to provide coverage. Once you have established linkages, and developed a framework that is sustainable and lightweight, then the emphasis on coverage is about working to fill in the gaps and weaknesses in your SR&ED process.
Of course, a framework feels like a fairly static thing – it’s not exactly a process, but more like the context for processes -- so thinking about or planning a framework, while it is important, is not going to be enough. So far, I’ve talked about the attributes or characteristics of a SR&ED framework – but what actually should make up such a framework?
I think there are at least three major aspects to a SR&ED framework, and maybe more, that should be embodied within processes and sub-processes.
First off, there needs to be a process – call it the Science process -- that surrounds the technical work itself … from the initial identification of potential SR&ED, to dealing with the actual SR&ED work, up to the detection of the end of SR&ED. That Science process needs to include the following sub-processes: detection (or recognition) and documentation. There also needs to be a second major process – call it the Financial Process – that concerns itself with demonstrating and quantifying the linkages between the Science effort and the costs, as well as satisfying some criteria around the nature of those costs. The third and final process revolves around filing the claim – as a part of your corporate taxes – and organizing your resources – material and human – to defend a SR&ED claim once it is made and is being subject to review.
Process thinking, for SR&ED, should be a cycle that strives continuously to achieve near-effortless and reflexive identification of potentially eligible work, followed or augmented by a relentless determination to first document and then demonstrate the linkages between the eligible work and its associated costs.
To be continued next week.
Bruce Madole
The Canada Revenue Agency (CRA) occasionally speaks about SR&ED claimants achieving a state which CRA calls “self-compliant”. On the surface, it seems like a reasonable goal – a claimant should have reached a level of sophistication, about the SR&ED program, which enables them to claim only what should be claimed, and presumably, never to claim what shouldn’t be.
It sounds like a happy situation, doesn’t it? No arguments with science reviewers or financial auditors about the eligibility or the scope of one’s claims. Perhaps, a level of frequent or at least open consultation to see if everybody still agrees with everybody else? Is this beginning to sound like “they all lived happily ever after”?
I’m in favour of self-compliance, as a goal, but there may be other goals that need to be recognized, such as the desire to maximize what is properly claimable. Does maximizing a claim sometimes entail crowding the edge of what is acceptable to claim? And if you do that, and the CRA pushes back, for science or financial reasons, have you stopped being self-compliant and become, suddenly, less trustworthy?
Myself, I don’t think trust really enters into it. I suspect that self-compliance is a matter of observable fact, and that it is probably also an historical observation, based on a pattern of claims filed over time, but without much predictive validity. (Unless or until, that is, a company reaches the level of process review with CRA, which does seem to imply a level of negotiated trust, not in the claims themselves, but in the processes that lead to them.) Perhaps, in an ideal sense, self-compliance means that we claim what we should, don’t claim what we shouldn’t, and then consult with CRA, in an open and receptive way, about the borderline bits?
This approach may result in a bit of under-claiming, from time to time, but you would have to think that such a working relationship would be healthier, in the long run, than ratcheting up the scrutiny, and even potentially triggering an audit, every time a claim is filed. So maybe self-compliance isn’t such a bad objective. I happen to think it’s a great idea, and more efficient, all round.
One thing, though. The law that governs SR&ED is defined by the House of Commons, as a part of the Income Tax Act. The final say about the interpretation of that law rests with the Tax Court of Canada (TCC); the CRA administers the law, but they don’t own it. The CRA itself has processes for unhappy claimants to follow in pursuing their concerns, which may ultimately arrive at the TCC. It has even happened, occasionally, that the TCC disagrees with CRA about CRA’s interpretation or administration of the law.
Therefore, it is important to remember that self-compliant, as a term, refers to being self-compliant with the law, and not merely self-compliant with respect to CRA guidance documents, policies, or expressed opinions, nor even, potentially, with the as-yet-unpublished claim review manual. The day may come when a claimant and the CRA disagree to the extent that the Tax Court of Canada must ultimately decide the matter: that is a claimant’s right, and the pursuit of self-compliance should not be seen or understood in any sense to alter that.
Bruce Madole
Some claims just aren’t that strong. Claims may be weak for a wide number of reasons – lack of clear technological uncertainty or technological advancements being sought, lack of evidence (or even a strong case) for experimental development, or even financial costs not commensurate with the core science. Whatever the weaknesses, it is also true that some of these marginal claims will have been prepared with the help of SR&ED consultants. Perhaps the consultants are just “taking their best shot” on behalf of a client. The question is: was this flawed or marginal claim submitted with the explicit expectation that smaller claims receive less scrutiny than larger claims by large corporations?
I believe it’s true that smaller claims have received less scrutiny in the past. I also believe that the Canada Revenue Agency (CRA) often pays greater attention to larger claims, on the grounds of “materiality”, which is accounting-speak for the fact that big numbers are often more important than little ones. There may be a degree of relativity in the way that the technical criteria have been weighed, between small to medium businesses (SMBs) and larger corporations. Large corporations can afford the most expert help, and they file large claims, and so the corporation is expected to demonstrate a more stringent standard than may be demanded of an SMB that is perhaps just getting introduced to the SR&ED program. I think there has always been a certain latitude or variability in the application of such standards. What is “routine” for one company may well be rocket science to another. The Science Reviewers try to sort it all out while trying to remain conscious of the fact that SR&ED is a tax-incentive program, designed to encourage technical innovation and the growth of technical know-how.
“Access” to the SR&ED program could easily become skewed heavily in favour of large corporations if such technical standards are too rigidly applied. On the other hand, how should we expect the CRA to behave if it were to become apparent that an entire segment of the SR&ED “industry” had grown up around the creation and filing of “marginal” claims, and with the explicit assumption that such claims would always fly under the radar? Might this not be perceived as an abuse of the latitudes previously granted? Might such an abuse not come to be perceived as “material” in its own way?
I think that such perceptions may well be a part of the CRA’s imminent introduction of a manual for SR&ED review. If that is a part of the unspoken story, and I believe it is, then small claimants (and the consultants who support them), should prepare themselves for a bumpy ride over the next few years. (The best approach to a truly marginal claim is probably to resist filing it in the first place, and that is a message for small and large claimants alike.) The push for greater granularity and more consistent scrutiny of claims will end up being more painful for smaller businesses than it is for large corporations, who can better absorb the overhead.
The entire SR&ED program hangs on a framework of inter-locking tolerances – technical and financial, human and legislative, legal and political. All of us who benefit from this program, and it is the best in the world, need to be sensitive to the rules and expectations under which we operate.
Bruce Madole
There’s an old non-joke about the recipe for elephant stew that read: “take one elephant – cube it.” Not all that funny – but we can imagine the effort involved in cutting up an elephant into little cubes of stew. It would be discouraging – about as discouraging as the effort to review one of those “mega-claims” the Canada Revenue Agency (CRA) used to get, before they introduced the short form.
No doubt some claimants have been persuaded, by some advisors, that some fairly marginal claims would sneak through the system because the amounts claimed are relatively small and “not material” when compared with many others. No doubt some claims may have been “puffed up” with extra words to give weight where real substance was lacking. However, I think that with the introduction of the short form, CRA has adopted a strategy that will introduce increased scrutiny for all claims, big and small, by leveling the size of individual claims.
The short form reduces things to a fundamental simplicity that leaves no room for the Cecil B. DeMille movie treatment – there’s not much room for smoke and mirrors in 1400 words. The CRA is has effectively forced us to “cube the elephant” for them – by forcing down the effective size of what can be described or claimed in one individual form.
That approach is somewhat punitive for claimants with larger claims – it drives up the costs of self-compliance – but it also means that, as an industry, the onus is now on us: we need to present claims in tiny, bite-sized packages that contain credible uncertainties, advancements, and work performed.
The implications of the CRA’s approach are clear: limiting the scope of what can be effectively claimed in a small space is the first step toward imposing a more-or-less standard level of granularity, so that more consistent review can be applied across the board, and more consistent (and perhaps more stringent) standards can be applied to every claim.
On the other hand, the use of the short form should make it apparent that a small business does not have to write a full length science-fiction novel in order to claim SR&ED – it’s now more like a technical (documentary) short story, about what you tried to do, and what happened as a result. Short and sweet – with a big return at the end of it, if you’ve got the facts right. That’s a process that could make the SR&ED program a lot more accessible to small and medium businesses, with or without professional help.
Seeking increased consistency, increased stringency, and increased access in a process of this magnitude is admirable -- achieving it would be miraculous. So perhaps we should manage our expectations and keep up the dialogue. This will all work out in the end.
Bruce Madole