When it launched in the early 2000s, global e-commerce platform Shopify was one of the first companies to offer small retailers an easy and affordable way to build an internet shopfront to sell directly to consumers. Started in Canada after its founders realized the software they had built to run a snowboard business was more valuable than the winter sports gear they were selling, Shopify went public in 2015, and today serves retailers of all sizes. In this episode of The Committed Innovator, Shopify president Harley Finkelstein speaks with McKinsey innovation leader Erik Roth about how he went from customer to company leader, how the company maintains its innovative spirit, and why he started a tea business during the pandemic. This is an edited transcript of their discussion. You can listen to the full episode on your preferred podcast platform.
Erik Roth: Harley, by every stretch of the imagination and criteria, you’re a true entrepreneur and innovator. Take us back to the beginning. How did you start creating companies?
Harley Finkelstein: I think there are two types of entrepreneurs generally. There are entrepreneurs by necessity, who basically have no choice but to start a business to survive. That’s my grandparents and my father. Then there are those for whom entrepreneurship is based in passion—there is something they really want to do, and entrepreneurship is the vehicle to do it.
When I was 13 years old, I went to a lot of bar and bat mitzvahs, and I thought the coolest people there were the DJs. They were like magicians to me, taking a crowd of 300 people sitting down eating a rubber chicken dinner, and within 30 seconds, the DJ would have a conga line going. I thought the magic DJs were able to create with music and their voice and energy was just the coolest thing I’d ever seen. So I really wanted to be a DJ, but at 13 years old, nobody would hire me. And I had this crazy idea to start my own DJ company and hire myself. That was my introduction to entrepreneurship as a tool to survive, but also a tool to solve problems like wanting to be a DJ and no one being willing to hire me.
Erik Roth: How did you make the leap from high-school entrepreneur to your first scaled company?
Harley Finkelstein: All throughout high school, I wasn’t really involved in the traditional extracurricular activities. This DJ company I had started was what I did on nights and weekends. I would find interesting new areas of business. I would buy videogames to sell to the bar and bat mitzvah clients. I would look for new people to hire so I could expand the business. I was tinkering all throughout high school with this DJ company. Ultimately, it became less about the craft of DJ-ing and more about the craft of entrepreneurship and business building.
When I finished high school, I moved to Montreal and went to McGill University. This was September 2001, and it wasn’t exactly the greatest time from an economic perspective. It was especially hard on my family. My parents lost everything, and I ended up starting my second business that year—a T-shirt company. But this time wasn’t about passion-based entrepreneurship. It was entrepreneurship based on necessity. I had to support my family. I had to pay tuition.
I started making T-shirts for universities across Canada. That was the first company I built that had any real scale. And I think that’s where entrepreneurship for me transitioned from being a cool thing that’s going to be a part of my life to my realizing, this is my life. The idea that entrepreneurship could save my life and save my family’s life—this is what I wanted to commit to. I ran that T-shirt business all throughout college and graduate school. I ended up going to law school, not to become a lawyer, but to become a better entrepreneur. A mentor of mine had this idea that law school would be like finishing school for entrepreneurship. And he was right.
But in law school I couldn’t skip class to run the T-shirt business the way I had as an undergraduate. I needed a business that could run on its own virtually while I was in class. The idea of being an online retailer was compelling. That’s when I met Tobi [Lütke], the founder of Shopify, and ended up becoming one of the first customers of that product.
Erik Roth: It sounds like your T-shirt business would have been exactly in the target market for Shopify back then.
Harley Finkelstein: It was. I met Tobi at a local start-up get-together in 2005. He was just transitioning from selling snowboards on the internet to selling the software he had built to sell those snowboards. A lot of people started asking him if they could use the software to sell their own products, including me. I was store 136 on the platform (Shopify now has millions of stores). I sold my T-shirts on Shopify all through law school and business school.
Erik Roth: Amazon, of course, already existed back then when you were building and scaling Shopify. How did you decide there was a market for a different type of e-commerce platform? How did you and Tobi go through that calculus of figuring out what Shopify needed to do and deliver in order to succeed?
Harley Finkelstein: There are two fundamental ways you can sell a product on the internet. The first is retail, direct to consumer, and the second is wholesale, selling through a marketplace or a third-party retailer. Remember, this was 2005, so at this point you had a couple of big marketplaces.
But those models effectively meant you had no direct relationship with your end consumer. You were really just renting customers from that marketplace. If your only goal is to sell as much volume as possible and you don’t care about having a relationship with the end consumer, you don’t care about a brand, you don’t care about building your own business with your own customer base, then that is a great option.
But many people want to have a direct relationship with their end consumer. They want to be able to sell directly to them. And they don’t want to rent customers; they want to own their customer base. They want to have all the information that comes along with that. Prior to Shopify, prior to 2005, the only way to create a beautiful, scalable, direct-to-consumer retailer online was to spend somewhere in the neighborhood of a million dollars hiring one of the large enterprises that would build you an online store.
The idea that a small business could actually build a beautiful, scalable online store that looks great, functions well, and sells to the end consumer—that was simply out of reach. So the reason Shopify was so compelling in 2005 was that it was the first time a small business was able to get access to the software and scalability that previously only the biggest companies could afford.
Erik Roth: Now you had a platform that allowed a business owner to create a relationship with their end customer in a way they could not at the volume marketplaces. How did you scale it, though, and get the economics and business model right?
Harley Finkelstein: Getting the basics right is very important. The fact that you can get something up and running very quickly that also can scale to many billions of dollars was critical. You can go to Shopify and in an hour or two you can set up a beautiful online store that looks uniquely connected to your brand, where the inventory functions well, where the checkout works well, and you get a merchant account automatically provisioned for you. I think fundamentally what gave us our start was there was real sophistication and robustness to the infrastructure of the Shopify product.
The way I can prove that to you is that some of the stores that got their start with us early on, in 2007, 2008, or even 2010, are not only still on Shopify today but are now multibillion-dollar companies. These are companies started at Mom’s kitchen table that are now the incumbents in their vertical, and they’ve scaled the entirety of their business life on Shopify.
We made sure there was no ceiling in terms of what stores could do on the platform. If they needed capital, we created a capital business. If they needed shipping or fulfillment, we helped them with that. If they needed to do something like payments or accept multiple pay methods, we enabled them to do that as well. If they wanted to sell in-store in a physical location, or they wanted to cross-sell on Instagram or TikTok or Google, everything they wanted to do, they could do with Shopify.
It allowed these brands to think of Shopify as a partner that is going to future-proof my business indefinitely in the long run. I think that’s the reason we built a lot of early trust with these small businesses. As they became big, we maintained that trust, which led to other existing large businesses eventually migrating to Shopify as well.
Erik Roth: The culture inside Shopify would need to be entrepreneurial and innovative to continue evolving the model to maintain it as a great home for these brands. Talk to us about the culture and how you created it.
Harley Finkelstein: Shopify is a company for entrepreneurs built by entrepreneurs. The vast majority of people at Shopify—we’re about 10,000 people today across many countries—are either entrepreneurs themselves or very deeply connected to entrepreneurship through their family or friends.
Erik Roth: Do you look for that as a hiring criteria?
Harley Finkelstein: Very much so. You can seek that out in a couple different ways. Obviously, if you have a Shopify business or you have a business to show us, that provides a lot of guidance on a candidate. It’s not because we’re thinking that if you started a business, you know how to integrate with payment gateways or you understand manufacturing, production, or supply chain. It’s because fundamentally there is something about entrepreneurship that is unique to the practice of building.
When you’re responsible for payroll, for the livelihoods of the people that work for you, it changes your entire DNA. It really does change the way you do things. Entrepreneurship fundamentally is all about resourcefulness. If you’ve gone through that journey of being an entrepreneur, you are inherently going to be resourceful, which means you are inherently going to bring that resourcefulness to Shopify. So we try to recruit and retain people who are deeply connected to entrepreneurs or are themselves entrepreneurs.
That’s the first aspect of our culture. The second is that Shopify is a place with ruthless prioritization. I’ll give you an example: we want people shipping, as opposed to meeting. So in January this year we did an exercise called Chaos Monkey, where we cleared everyone’s calendars right after the year-end holidays to test whether we really needed all the meetings that were happening.1 People came back from holiday and their calendars were empty. I think we deleted something like 12,000 recurring meetings. If you quantify that over an hourly basis, we deleted 322,000 hours of meetings. That is 36 years of meetings. That’s not to say that some of those meetings were not important. Some may have been. But we had a cool-off period before people were able to add those meetings back.
We always try to think about ways for us to create something like altitude training for runners. The Chaos Monkey activity was altitude training. Most of the meetings we deleted were not added back, which suggests that they weren’t that important. Why does that matter? Because we try to bring in people who are deeply entrepreneurial and want to work at a place where they have this ruthless prioritization, where they think about subtraction and want to be really innovative and able to pivot without their egos getting in the way.
Erik Roth: That’s an area many organizations struggle with. Sometimes we even put in place a metric where we track how many times decision makers say no. How do you help people prioritize by being able to say no?
Harley Finkelstein: When you subtract or reduce scope and say no to something, you provide yourself with an opportunity to find a better idea. But every time you say yes to something, you are basically locking out every other idea that may come to the table. There are times where we have a great idea and we think, “OK, this is really going to add a lot of value.” For Shopify, though, unless that idea unlocks true value for most merchants most of the time, it’s probably not something we’re going to do.
To supplement that, we have a massive ecosystem of third-party app developers. The Shopify App Store is probably one of the most robust developer ecosystems on the planet. We have about 8,000 apps in our app store. That is the perfect place for third parties to build functionality that some Shopify merchants need most of the time, or most need some of the time. But if most people need something most of the time, we’re going to build it ourselves because we believe it has to be world-class and out-of-the-box perfect.
Here’s an example of that prioritization on the go-to-market side. When those Shopify stores I mentioned that started small with us scaled to become some of the largest brands on the planet, there was this inherent desire for us to move upmarket—if we can serve those companies, we should go and get much larger companies to use Shopify.
But at that point we felt we weren’t ready for that, partly because we didn’t at that time understand the enterprise market really, really well—the way they buy, the way they do procurement, and integrating with their existing ERP and inventory systems. So we did a deeper dive on how these massive brands work at scale, then once we felt we were ready to compete for enterprise business, and had done the work to deserve to win their business, we went out to bring on these massive brick-and-mortar brands.
A lot of companies do that the reverse way. They go to market first, but the product is just not ready. Maybe a criticism of Shopify might be that sometimes we overbuild things. But because the business we are in effectively runs the entire business for our merchants, for our customers, we take that approach. If your blog platform goes down, that’s unfortunate, but you’re still going to live to see another day. But if your retail operating system, which is the role we play in the lives of these millions of stores, goes down, you don’t make money. Your business is going to be in jeopardy. We take that responsibility very, very seriously.
Erik Roth: That sounds like a bank—ATMs must transact, payments must process. Often that need to deliver all the time without fail can stifle innovation, though. How do you make sure you’re making the right calls in optimizing for what is valuable to everybody? How do you make sure you’re not pushing out some of those seemingly small ideas, which could be the next big ideas, and just defending the core?
Harley Finkelstein: Here’s something most people don’t understand about Shopify, but I think it speaks volumes. If you were to pretend that Shopify was a single retailer, if you were to aggregate all our stores in the US alone, for example, we would be the second largest online retailer in America after Amazon. That’s important because if you are the second largest retailer in America, you are entitled to massive economies of scale—shipping rates, credit card transaction rates, or rates on borrowing of capital, for example.
Because we’re the platform for these millions of stores, we’re able to disseminate those economies of scale to all of these small businesses. When you are a small business just getting started on Shopify, you are actually getting rates on these costs that previously only the biggest companies on the planet were able to get.
What we’re trying to do for the smaller brands is level the playing field so they have a better chance of success. At the same time, once those brands get really big, or larger established brands come to us for their e-commerce needs, we also need to figure out what they need to grow as well. Sometimes both types of companies benefit from similar capabilities. For example, even though you may be running massive businesses and brands at scale, you also want to be entrepreneurial. So when you hear that Instagram now allows commerce to happen right on the platform, or you see things like live shopping on YouTube, or you see that TikTok is now embedding a checkout button, you want to be able to take advantage of that.
Our job is to continually requalify for the right to be the retail operating system for all these brands; whatever they want to do, the answer is yes with Shopify.
Our job is to continually requalify for the right to be the retail operating system for all these brands; whatever they want to do, the answer is yes with Shopify. Either we do it ourselves because we think it’s that important, or we leverage our APIs and our third-party ecosystem. There should never be something you want to do in commerce or retail on Shopify where we say, “That’s impossible.”
We talked about this in the sense of omnichannel sales on our IPO roadshow in 2015, and the fact that brands are going to want to sell across every surface area where a consumer exists. We saw a lot of eye-rolling at the time, with people saying, “Yeah, yeah, you guys are just e-commerce.” But today if you look at all these amazing brands, almost all of them, to a tee, are multichannel. None of them are single channel. None of them only sell online, or only sell in physical stores. They sell everywhere their consumers are.
What we’re trying to do is not only make sure that the things you want to do right now on Shopify are available and robust and world-class, but also that the things you may want to do years from now are also available, whether it’s AI-based commerce, or Web3, or augmented reality. We’re working on those things now, not because people want them right now, but because we anticipate in the future they may. And if we think in the future they may, we have to make small bets now in order to qualify for that later.
Erik Roth: Should we stop talking about e-commerce? Is it just the wrong word? Maybe it doesn’t matter anymore.
Harley Finkelstein: I haven’t referred to Shopify as an e-commerce company in probably four or five years. I believe the brands that will be most successful in the future will be entirely channel-agnostic. They’re going to be focused on how to serve their customer best.
Erik Roth: So when you see a TikTok with a checkout button, do you get nervous about that?
Harley Finkelstein: Shopify is the commerce partner for all those platforms. Any new platform that pops up or any new functionality that pops up within a platform—we’ll build an integration for it. We’ve been integrated with Google Shopping and YouTube for many years, and when you see live shopping inside of Google, people posting products underneath the video on YouTube, that’s all powered by Shopify.
The key for us is to make sure we are the first call that all of these surface areas make when they want to get into commerce. And more and more, every major brand that all these social media platforms and surface areas want to have integrated with them is a Shopify store.
Erik Roth: To make all those small bets, though, and do everything you just described, there have to be little pockets of activity within Shopify that are already working on those areas. How do you nurture the small bet and make it relevant and protect it so it doesn’t get killed in the wake of something more important for the core?
Harley Finkelstein: Sometimes a leader has to put their proverbial arms around a little sapling tree and say, “I’m protecting this tree because one day it may grow.” But that goes back to the conversation earlier, Erik, around having entrepreneurial people. They’re deeply passionate. Let’s just use AI as an example. Everyone’s talking about AI now, but we’ve been experimenting with it for a very long time for product descriptions, knowing that at some point AI is going to get so good that all you will need to do is take a photograph of a product and the algorithm will write you the most amazing, highly converting product description out there.
Another example is augmented reality (AR). We have a small team of people working on using it for selling things like furniture, where you can use AR to place furniture around your home virtually to see how it looks before you buy.
Another example is Web3. Last year it was a big deal; now maybe it’s less spoken about. But for us, there are components of all of these massive trends that we think are incredibly valuable for commerce. With Web3, token-gated commerce is super-compelling—you could use a non-fungible token (NFT) for access to events and commerce within events.
Ultimately, what we’re trying to do is mine all this new technology to figure out how it might work. For now, they are very small teams with limitations. If those projects hit certain thresholds, then we’ll invest more. But for a long time we have had these tiny, almost little start-ups inside of Shopify working on such potential projects.
Erik Roth: How do you actually create the organizational model within to enable everything you just described?
Harley Finkelstein: I think way too many companies overly rely on the org chart. For example, I have teams that report to me that are three-person teams. They don’t necessarily fit neatly into the org chart, and that’s OK. One thing I think a lot about is there’s a deep difference between founder-led companies and professionally managed companies. Shopify is very much a founder-led company. It allows you to say, “I don’t really care how this fits into the org chart. I’m not going to let the org chart dictate innovation.”
I know everyone talks about mission-driven and culture and all that, but fundamentally you actually have to believe. You can’t put leadership posters on the wall and expect people are going to find that compelling. At Shopify, we believe that more entrepreneurs mean better things for the world. More entrepreneurs mean more survival, more success, more independence, more problem solving. So if you believe that entrepreneurship is important, Shopify is a great place to work.
Erik Roth: Speaking of founder-led companies, you are a founder once again. Talk about why you stated Firebelly Tea. Why now? You don’t have enough to do running Shopify?
Harley Finkelstein: The truth is it’s really fun. But I can tell you a bunch of other reasons why. During the pandemic I was drinking way too much coffee and my anxiety was peaking. I needed an alternative. My best friend is a tea aficionado and he said, “Hey, try this amazing green tea.” Turns out I’d never had truly high-end green tea. Then I started giving it to other friends, and they were amazed by it, too. Now I have an entire group of friends and family members who don’t drink coffee after noon—they drink green tea. Everyone seems to be operating at a super-high level, and anxiety levels are reduced. Part of it was this discovery that, like me, most people have never had truly incredible green tea before—matcha, sencha, there’s a bunch of other Japanese greens that we love.
Another reason was that the last time I started a business was in 2006 or so, and that was my Shopify store. Things were really different back then. The growth funnel in 2006 for an online store was fundamentally just AdWords. That’s all it was. The more you spent, the more you got. Today it’s much more complex. Content marketing is important. Influencer marketing is important. Social media is important. The brand having its own persona, almost like it’s a human in itself, matters more today.
I felt I was a little bit out of touch with how to build a modern business. So my best buddy and I launched Firebelly Tea during the pandemic. Not only are we learning, but we’re testing everything on Shopify. I’m testing Shopify audiences, fulfillment, integrations, and social media platforms. For us, Instagram works well. TikTok doesn’t work as well some weeks, but others, when we put out a physical item like a travel mug for the tea, TikTok works great. It’s so fascinating. While some people like to watch football on Sundays, I’m tinkering on little businesses with my friends.
Erik Roth: Harley, this has been amazing. Thank you so much for joining us today.
Harley Finkelstein: Thank you, Erik.