Tell us a little bit about your career. What was your path to becoming a VC and founding N49P?
I didn’t follow the typical venture career path of completing an apprenticeship and then launching a fund. Instead, I got my start in consulting and investment banking, where I spent years focused on tech M&A. Those experiences allowed me to start building my network and taught me how to communicate effectively to different stakeholders. After that, I became an operator, working as an early employee at three companies before co-founding my own business. That phase of my career gave me a lot of experience getting companies to product-market fit.
In 2014, I started TechTO and subsequently helped bring AngelList to Canada. Around that time I also became an angel investor. Collectively, all of these different experiences allowed me to develop a really good view of the Canadian tech ecosystem, positioned me well within it, and helped me understand how to structure successful deals. In 2018, I decided to start my own fund and launched N49P with my partner.
N49P is kind of a unique name. What’s the significance?
It stands for north of the 49th parallel and is a reference to our Canadian focus. We partner exclusively with Canadian startups or founders, including those based overseas but that have a significant presence here.
What else can you tell us about your fund?
We invest in early-stage companies and want to be the first institutional check that they take. And, unlike a lot of early-stage funds that take more of an index-like approach by investing in between 50 and 100 companies per fund, we only make 25 to 30 investments in ours.
One reason why we invest in fewer startups is because it allows us to work more closely with our portfolio companies. That’s important because we know that no matter how much founders raise or how good they are, there are a bunch of key inflection points that they’ll all encounter during their first three years on the job. One of our areas of focus is bringing our founders the expertise they need to remove as many of those roadblocks as possible so that they can focus on growing their business.
It’s also worth pointing out that if you look at our portfolio, you’ll find quite a few companies that most other VCs probably wouldn’t have funded. That reflects the fact that we are more willing to invest in different founders and unproven markets than most Canadian funds, provided we believe in their vision.
You mentioned being early-stage investors, which is somewhat of a moving target these days. What does early-stage mean for N49P?
It’s a fair question given how much things are evolving. We’ve invested in everything from small, nascent teams with just an idea to more established companies generating as much as $40,000 in monthly revenue. What matters most to us is finding great teams with high conviction in their unique insight about the future that has the potential to create a massive opportunity.
Can you tell us about some of your recent investments?
There have been a handful lately, so I’ll just highlight a couple. One that comes to mind is EZee Assist, an all-in-one AI support assistant that helps franchisers reduce the amount of time they spend supporting their franchisees. It does so by using generative AI to help franchisees get the information they need, while freeing up the franchiser to focus on strategy. We made the investment because the company has an incredibly talented team and because they have a clear vision about where their product can go and how it can help multi-location businesses.
Another company we recently invested in is Reputable. It’s a Waterloo-based business that helps its customers validate their product claims using advanced wearables and AI. Their platform cuts clinical trial costs while delivering a lot of data-driven insights, and it does so a lot faster than traditional studies. We like the company for a variety of reasons, not least because the founder has an incredible track record of building companies and the considerable tailwinds propelling the health and wellness industry.
As artificial intelligence proliferates, entrepreneurs face a unique combination of challenges and opportunities. What kinds of conversations are you having with founders to help set them up for success in this new era?
We spend a lot talking about how SaaS is evolving in the age of AI and, more specifically, what that means for their pricing models.
Anyone over the age of 40 can probably remember a time when software was sold in boxes, often in conjunction with some kind of customer support contract. It was an upfront cost that came with the expectation that customers would periodically want to buy newer and better versions of the software as it became available. Of course, SaaS changed all of that. It was delivered online and was constantly being updated, meaning that customers always had the latest and best version of the product. And rather than pay upfront as they used to, it was sold on a subscription basis with the price determined by volume or number or seats.
As SaaS became more sophisticated over the past 10 to 15 years, the model shifted again. Rather than charge by the seat, which could be messy if seats went unused, SaaS companies increasingly tied their pricing to the value they delivered. That value was typically measured in usage, such as the number of people actively using the product, the number of records being created in a CRM, or the number of emails sent in a marketing automation platform.
Now with AI in the mix, the model is shifting again.
How so?
Using AI has some important implications. First, it has a much higher compute cost, which can lead to lower margins than software providers have traditionally enjoyed. Then there’s the fact that rather than simply digitizing a process such as managing customer relationships in the case of a CRM, a lot of AI produces actual outputs. AI-powered legal software, for example, might produce demand letters. That, in turn, reduces the number of paralegals a law firm would otherwise need to do that work.
This combination of higher costs and a shift from digitizing processes to delivering outputs is important. Since AI reduces the number of people needed to complete a specific task, charging by seat no longer makes sense because the software is literally reducing the number of seats required to complete the work. As a result, AI companies are experimenting with different ways to charge customers. In the law firm example I just mentioned, it could be per demand letter produced. That makes sense because as a law firm grows, the number of letters it needs to write will grow, allowing the software provider to charge more.
That sounds pretty straightforward.
Sure, but it’s not the only scenario. Imagine a software company whose pricing model is based on the number of customer service queries being answered. If the software is any good, over time it will likely reveal patterns about where questions routinely arise. Those insights could be used to make whatever changes are necessary, say to a company’s website, to clarify a particular issue, resulting in fewer customer service queries. So not only is AI reducing the number of employees required to do a job, it’s probably also reducing the amount of a given activity that can be charged for.
So if AI makes a company more effective, and that affects the fee the software provider can charge, the question becomes how do you capture some of that value?
Is it by charging a percentage of what it would cost to have an employee or a third party do the same work? There’s a lot of experimentation happening right now to figure out what works best.
So it sounds like we may see multiple new pricing models emerge with AI?
It’s all pretty messy right now. Ultimately, I think how fees align with outputs will vary by category. There will be different pricing models for companies that help you generate revenue versus those that help you become more efficient and effective. It may well end up varying on a case-by-case basis.
In light of all of this, what advice are you giving to founders to help them navigate this shift in SaaS pricing models?
I remind them that the underlying technology is changing incredibly quickly. So it’s really important to remain flexible because what’s working now might now work a year or two. Practically speaking, that means that rather than develop a pricing model based on what’s working now, think about where your platform is going and keep experimenting and trying to optimize as things continue to evolve.