The e-commerce platform space is in a race to acquire merchants. It’s a long ride. Most of the competitors are riding the Cervelo R2. It’s a good road bike; carbon frame, decent components, and affordable. Some competitors have mounted up with the S3. The seat is raised and gives them more power, they have deeper rims with better aerodynamics, and some have even added aero bars. The S3 riders are clearly pulling ahead from the R2 pack. And then there is Shopify, who has mounted up with the P5. It’s the lightest bike in the race, outfitted with electronic shifters to make shifting seamless. It has the deepest wheels, positions the rider for maximum power, and has disc brakes. All the bells and whistles. Leading up to the race Shopify geared up with a power meter, the Fenix 6, Zwift, and the Wahoo Kickr. They trained the hardest, and have the best ride. At the start of the race, Shopify used to look back over their shoulder and see the pack of S3 riders hard on their tail. But now… now they look back and see no one. At the turnaround, Shopify sees that the S3 riders haven’t given up, but barring a flat tire, Shopify isn’t worried. Riders that missed the starting gun have all but given up. At this point in the race, spectators seem to agree that the Canadian rider on the sleek P5 is going to take it.
Investment Thesis
When we first wrote about Shopify (SHOP 0.00%↑) it was trading at US$475/share. Using the outside view, we had concluded that market expectations were extraordinary, and would be hard to meet, let alone beat. Despite the share price increasing >70% since then, we are now beating to a slightly different drum. We realize that our reference class and time horizon were both inappropriate, and that our net margin assumptions were too conservative. Conversations with other investors helped us better understand the industry and company-specific KPIs, and why the outlook is so positive. We’ve come around to the idea that Shopify really is an outlier in public equity markets.
Our research suggests that global e-commerce penetration could triple in the next 10-15 years, and that e-commerce platforms will be more central to that growth than in the past. We expect that most merchants will want a single outsourced platform from which to manage all facets of their business; a platform that allows merchants to go directly to consumers, but also to utilize online marketplaces and other distribution channels. In our view, Shopify is the clear leader in the e-commerce platform business, and should gain share through our forecast period. Not only should Shopify capture a greater share of subscribers, but they are also well positioned to offer ancillary services to that subscriber base, like payments, fulfillment, and banking. Many of these ancillary services are hard to offer without scale, which differentiates Shopify from other platform providers, and contributes to the positive feedback loop of subscriber growth. In addition, these services result in sticky customers, who might eventually be less price sensitive on the core subscription product. The roll out and monetization of ancillary services is still early, and we expect that it could result in a 3.0x increase in Shopify’s average 2019 take-rate. The combination of greater e-commerce penetration, more outsourced websites, and market share gains, could result in GMV growth of more than 20x over the next fifteen years. Our base case suggests that GMV growth and higher take-rates should translate into revenue growth of >30%/year through the 15-year forecast period.
It’s really difficult to know where our expectations differ from the market (if at all), but we suspect that we have a modestly more positive outlook on three components of the story: first, we think SFN penetration and profitability will ultimately exceed expectations; second, we don’t think that the market expects 40% global e-commerce penetration of retail sales in the terminal year; and third, we doubt that the market assumes the same magnitude of market share gains as our base case. Bringing our estimates down slightly in each of these three categories simultaneously is sufficient to arrive at the current share price.
Our base case implies 17% upside to the current price of ~US$800/share, but shouldn’t be viewed in isolation given the wide range of potential outcomes. The return to our Bear and Bull case is -59% and 188% respectively. Despite spending a lot of time thinking about the KPIs that drive these outputs, it’s pretty clear that the standard error on our forecasts is going to be huge. We are guaranteed to be wrong about multiple KPIs. Nevertheless, if we were to draw a distribution of outcomes, that distribution would have a pretty fat right tail, and we think that the current share price is somewhere left of the 40th percentile. There will probably be a more attractive price to own Shopify, and even though the asymmetric risk profile is compelling, we hold the view that it’s difficult to make a bet with high conviction today. At US$500-600/share, this would be a big bet for us. At US$1,300+/share, we wouldn’t own any. If you happen to have a differentiated view in either direction that you’d like to share, we encourage you to reach out.
Value Proposition
The value proposition of e-commerce platform businesses is pretty clear: help merchants tap the growing e-commerce pie, provide them with tools to manage their business more efficiently, and give them more control over their brand, the customer relationship, and digital distribution channels. Once merchants move online, they don’t go back, so this is a need-to-have service. Shopify charges merchants a monthly subscription fee to provide this critical infrastructure, and the fee tends to be a very small share of wallet.
In addition to the core e-commerce platform, Shopify provides a number of ancillary services that they bucket into the Merchant Solutions business. These services include payment processing, payment hardware, shipping and fulfillment, and lending. In each of these verticals Shopify aims to offer a best-in-class service at a best-in-class rate, and they do this by aggregating volumes to drive down unit costs. By providing all these ancillary services, Shopify is becoming a one-stop shop for everything merchants need to run an online business. The combination of convenience and efficiency should drive adoption by merchants.
We expand on the specifics of each business below.
Subscription Solutions
Peeling back the onion
The core Shopify subscription business has two key drivers: number of subscribers, and subscription fee per subscriber. The company offers tiered subscription packages to accommodate everyone from small entrepreneurs to massive enterprises. More specifically, they offer: Basic for $29.99/month, which targets small entrepreneurs; Shopify for $79.99/month, which should capture most small and medium sized businesses (SMBs) with at least a few employees; Advanced for $299.99/month; and Shopify Plus, which targets larger merchants with complicated SKUs and ranges from $2,000-40,000/month based on GMV. Shopify Plus customers include companies like Staples, Heineken, General Mills, Heinz, and Cirque Du Soleil, to name a few.
Shopify doesn’t explicitly break out how many merchants subscribe to each service, but we’ve attempted to back into the historical split as shown in Exhibit A. Even if we assume a large standard error on our estimate, it’s clear that the Basic plan is the most popular, which means that the majority of Shopify’s merchants are small businesses. From a revenue perspective the story is a bit different, with only half of 2019 revenue coming from Basic subscribers. It’s clear that merchants who grow on the platform become exponentially more valuable customers to Shopify. It’s impressive that Shopify has grown the subscriber base at a 5-year 50% CAGR and subscription revenue closer to 60%, with effectively no change in subscription fee per tier (with one small exception that we’ll touch on later).
A common misconception
A common bear thesis on Shopify is that churn in the Basic bucket must be massive given the high failure rate in that cohort, so Shopify will need to constantly reacquire customers. This makes it difficult to realize margin expansion as the business grows. However, if we look at subscription revenue per tier instead of total subscribers, we see a much different story. A growing portion of revenue appears to be generated from the more expensive tiers, which means larger businesses with lower failure rates. As a starting counterpoint, revenue churn is much lower than customer churn.
The second thing to consider is that some businesses will actually move up into more expensive subscription tiers. The price of a Shopify plan is 2.7x the price of the Basic, while the starting price for Shopify Plus is 67x the price of Basic. If a single Basic customer grew into a Shopify Plus plan, it would completely offset the lost revenue from 66 small merchants going out of business.
To more accurately assess what revenue churn might look like for just the Basic bucket, we pulled U.S. Bureau of Labor data on SMB failure rates from 1994 to 2018 (Exhibit B). If the average Basic subscriber was 2 years old today, then 55% would likely go out of business over the next ten years. If half of the surviving merchants ended up upgraded their plans from Basic to Shopify, then the annual revenue churn over the ten-year period would end up being only 2%/year. In our view, this is actually quite low. This also ignores the fact that surviving merchants end up paying for additional services as they grow.
How big is this industry, and how big can it get?
Shopify estimates that there are 47 million SMBs globally (less than 500 employees), and our research suggests that only a fraction of them are using one of the big e-commerce platform providers. We used BuiltWith to add up the number of websites powered by the seven biggest non-Chinese platforms in the market, and we see that they cumulatively support only 9% of global SMBs; including our rough estimate for China, only 13% of global SMBs are using one of these platforms (Exhibit C). It’s difficult to get a time series of this data, but 5 years ago we estimate that less than 2% of merchants would have been using one of these platform providers. Clearly the market for outsourced e-commerce platforms is growing quickly, and is still meaningfully underpenetrated. There is probably some upper limit on how many SMBs actually need an e-commerce platform, but once again, it’s difficult to pinpoint where that ceiling is. Our 90% confidence interval is 50-70% of SMBs. If we take the mid-point, then these platform providers could see SMB subscriber growth of nearly 5.0x from 1Q20. Over the last 30 years, the number of SMBs in the United States has also grown at about 60 bps/year, and we’d expect to see this pie continue to grow at a similar rate moving forward.
We also estimate that there must be a couple hundred thousand enterprise businesses globally. Many of these enterprises don’t have a DTC platform, while others have built their own e-commerce infrastructure. In both cases, we think the benefits of outsourcing this aspect of the business creates value for merchants. Jeff Silverman, the Head of Global Ecommerce at Steve Madden, summed it up perfectly when he said:
“We didn’t want to be in the technology business. We wanted to worry about selling shoes and be great at what we’re great at. That’s what Shopify Plus empowers us to do.”
What does the competitive landscape and market share look like?
The data from BuiltWith shows that WooCommerce has the largest share of total subscribers, but falls to second place among the top 100k sites. We note that WooCommerce is just a plug-in for WordPress sites, and aren’t convinced that this data accurately reflects meaningful e-commerce intent. In our view, Shopify is leading the charge in real subscriber market share, powering nearly a third of all identifiable e-commerce websites. Ignoring our view of the data, it’s clear that WooCommerce is Shopify’s largest competitor in the small business space, while Magento is the largest competitor for enterprise (competes with Shopify Plus).
It’s incredibly difficult to get a time series of market share that we can comfortably rely on, so we turned to Google Trends search interest as a proxy for market share change through May, 2020 (Exhibit E). We note that only about 20% of Wix subscribers fall under e-commerce, so that specific search word is probably misleading in this context. Nevertheless, it’s clear that Shopify is the most popular provider, and that this relative popularity is becoming more pronounced with time. This data fits the narrative we’ve heard, which is that Shopify is taking market share from other providers in the space.
We spent some time looking at each of Shopify’s competitors, and one thing stood out to us immediately: few competitors, if any, are “merchant obsessed”. Most competitors, including Wix, Squarespace, Weebly, WooCommerce, and Magento, are either part of larger entities with other products to focus on, or aren’t exclusively focused on e-commerce: the majority of Wix, Squarespace, and Weebly subscribers seem to be outside of e-commerce; WooCommerce is actually just a plug-in for WordPress websites, which makes onboarding awkward; and Magento is owned by Adobe, but has changed hands multiple times in the last decade, and only focuses on enterprise customers. Shopify, on the other hand, has been a standalone entity since 2004, and only focuses on one thing: e-commerce. As a result, Shopify seems to have a better product, with faster updates, a simpler onboarding process, and the widest offering of solutions.
In lots of cases, competitors don’t even target the entire market. WooCommerce doesn’t have an enterprise product, and Magento doesn’t have an SMB product. The team at WooCommerce won’t be thinking as intimately about the commerce problems that enterprises face, and the team at Magento won’t be thinking as intimately about the problems SMBs face. Without the holistic view on product development, and the same economies of scale benefits on R&D and marketing, it will be hard for WooCommerce and Magento to effectively compete with Shopify.
As we learn about Shopify’s culture, it’s clear that they are singularly focused on helping merchants. They want to be all things to all merchants, and Tobi, the founder and CEO, has repeatedly highlighted that Shopify is “merchant obsessed”. It reminds us a lot of Amazon’s focus on the customer:
“If there is one thing Amazon.com is about, it’s obsessive attention to the customer experience. End-to-end.” – Jeff Bezos, 1999
In our view, this obsessiveness has led to a better platform, and explains the historical shift in market share. We read a ton of third-party product reviews, which compared Shopify to other platforms, and genuinely laughed out loud at this one:
“Shopify is like Samwise Gamgee. It has one purpose in life: To help Frodo destroy the ring. Shopify’s one purpose in this world is to help people create and maintain an e-commerce store.
Wix is like Merry and Pippin. They definitely want to help Frodo destroy the ring, but they have other stuff going on. There are fireworks and ale out in the world. Also, something about a giant tree-man.”
In our evaluation of the competitive landscape, we also noticed that Amazon and eBay tried (and failed) to create their own e-commerce platform business. eBay purchased Magento in 2011, which became a major part of their e-commerce business. Within four years, an activist-led restructuring ended with Magento spinning out to a private equity business. Our view is that Magento was non-core to eBay, not receiving the appropriate TLC, and might have even cannibalized part of the marketplace business (low conviction view). In 2018, Adobe bought Magento, which marked the fourth owner in just ten years. Software is all about people, and the quality and motivations of people are influenced by culture. No company culture can remain robust through a decade of uncertainty and shifting ownership. We have no doubt that Magento was neglected at eBay, and suspect there is a good deal of technical debt within the platform today. We hear that Magento is a much clunkier product than Shopify Plus, and takes much more time to onboard and maintain. As a result, it appears that lots of Magento customers have left the platform for Shopify. For example, when Magento was spun out of eBay in 2015, Forbes wrote an article highlighting the fact that Magento powered online stores for big merchants like Rebecca Minkoff (link). Within three years, Rebecca Minkoff had migrated to Shopify Plus (link), and touted the fast on-boarding process and superior product that Shopify offered. It’s an interesting saga, and is a good example of an established marketplace business failing to run an e-commerce platform business. It also highlights that Magento is probably in a weak relative competitive position today.
Amazon approached the e-commerce platform business organically. They launched the Webstore in 2006, but by 2015 decided to wind it down. It’s not totally clear why Amazon failed, but one common narrative is that the Webstore had too much “amazonishness”. The Amazon logo was always visible, the product wasn’t fantastic, and the merchant had little control over their own brand. Amazon’s success in everything else is in large part due to their extreme consumer-first attitude, often at the expense of other stakeholders, including merchants. In hindsight, it’s pretty obvious that a company who had always put the consumer first would have a difficult time building an excellent product for merchants. It just wasn’t in their DNA. What’s wild is that Amazon actually referred all their Webstore clients to Shopify (link). There is no greater kudo’s than competing head-to-head with Amazon, winning, and then having the e-commerce behemoth single you out as the best platform. Now Amazon is just one of the many distribution channels that Shopify merchants can access.
For both Amazon and eBay, we notice that the e-commerce platform product could very likely have cannibalized the core business. Empowering merchants to go direct-to-consumer and use other distribution channels almost certainly creates some direct competition with the core marketplace business. Hamilton Helmer writes about the idea of counter-positioning in his book 7 Powers, and while we’re new to the concept, in many ways it looks like the e-commerce platform model is a form of counter-positioning relative to traditional e-commerce marketplaces. In our view, this will probably keep the incumbent marketplace operators from successfully competing with the likes of Shopify.
We’re also skeptical that any of the other marketplaces – like Facebook Shops or Google Shopping – would attempt to build competing platforms. The boat has likely sailed to build a similar quality product without risking a large NPV-negative outcome. Rather, we suspect merchants will continue to rely exclusively on the incumbent e-commerce platform providers.
Shopify App Store
Many of the e-commerce platform providers have their own app store, which allows third parties to develop apps for merchants that are easy to tie-in to the platform infrastructure. Shopify is no different in that regard. Many of the platform providers also develop their own apps, but this is where we think Shopify might pull ahead of the pack. They appear to have more internally developed apps, and they seem to be more powerful for merchants. For example, Weebly has developed 13 of their own apps, whereas Shopify is at 30, many of which were rolled out in the last few months. Apps like Shopify Flow allow merchants to automate workflow, like automatically reordering inventory when supply gets low. We haven’t seen tools like this in other marketplaces (although we may have missed them). A lot of this comes down to R&D, and as the largest platform provider, Shopify can afford to accelerate spending on apps in a way that some of their competitors might struggle to do. This probably contributes to the flywheel, where the additional tools create greater appeal for the platform, which helps Shopify collect more subscribers, which allows them to increase spending on new tools, and repeat. It’s not a big part of the business today, but is certainly a differentiator at the margin.
Share of the customer wallet
The US Small Business Administration reports annual revenue for small businesses by size, and that data suggests that subscription fees for the majority of Shopify merchants are well below 1% of total revenue. Exhibit F shows subscription fees as a share of net income for merchants if we assume a fixed 10% net margin across all business sizes.
All of the data we’ve come across suggests that merchant profits increase by substantially more than the cost of subscriptions once they create an e-commerce platform. For example, our analysis suggests that the average retailer with 5-9 employees sees their profit pool grow by 15-20x the cost of a Shopify subscription. The e-commerce subscription service is also increasingly a need-to-have for most retailers, particularly as e-commerce penetration grows. It’s akin to four walls and a roof for a pre-internet retailer. Once a merchant goes digital, they don’t go back.
Given how important and inexpensive the service is, it’s possible that Shopify could increase prices in some categories. One way they could do this is to tie subscription fees to GMV for the Shopify and Advanced plans, much like they do with Shopify Plus. With a high enough threshold, perhaps only merchants with 20-99 customers end up paying more. Shopify’s share of wallet in that category is so small that we doubt a merchant would even notice a 25% price increase over a three-year period, let alone care.
That being said, in a subscription-solution-vacuum, where lots of competitors have a product that’s not drastically different, it’s unlikely that Shopify could substantially increase pricing without impacting churn. However, Shopify continues to layer in other value-added services that could become increasingly unique to the Shopify platform, like fulfillment. We’ll expand on these next, but if these services remain unique to Shopify, we think it’s likely that at some point in the future the company might be able to exert some pricing power on subscribers. Shopify’s average subscription take-rate on merchant e-commerce GMV is about 1% today, which compares to the 13% that Amazon charges merchants for products sold through their marketplace. In our opinion, Amazon doesn’t add 13x more value to merchants than the core Shopify platform (although it’s not a perfect comparison). In any event, we probably wouldn’t pay for pricing power today, but it’s important to keep in mind.
Geography
The vast majority of Shopify’s revenue comes from merchants in the United States, which probably isn’t surprising (Exhibit G). The opportunity for growth in the United States is still massive, and it will probably be the biggest driver of revenue growth in absolute dollar terms over the medium term. That being said, the global opportunity set is probably orders of magnitude larger, and presents a massive opportunity to Shopify. Exhibit G also shows year-over-year growth rates by geography, and it’s worth highlighting two things: first, even though revenue growth is decelerating, it’s still astronomical; and second, growth rates for ROW have been notably higher than other geographies in recent years. As Shopify rolls out their platform to more countries, and increases the number of services available in those countries, the ROW bucket is likely to continue growing much faster than the other four geographies that Shopify breaks out. We note that China seems to be an elusive market for most non-Chinese businesses, and probably stays that way for Shopify. But, if Shopify is successful in non-Chinese international markets, we suspect the ROW could make up a significant portion of revenue in 10 or 15 years.
Merchant Solutions
Once merchants come onto the Shopify platform, there are a number of additional services they can choose to use. For the most part, monetization of these services is tied directly to total GMV flowing through that channel. The two most visible channels are payments and fulfillment, but a number of other businesses are also being developed. A lot of these incremental products and services are just that: incremental. But, when combined they start to differentiate the platform from other competitors. They also increase the barriers for new entrants by increasing the upfront costs of developing a fully comparable service. Before we dive into the services, we’ll explore e-commerce GMV.
E-Commerce GMV
Growth in e-commerce is one of the biggest secular themes of the last decade, and we think it’s a theme that will continue to be relevant for a very long time. Exhibit H shows estimated e-commerce penetration in the US and China versus the global average. It’s interesting to note that e-commerce penetration in China is orders of magnitude larger than almost anywhere else in the world, and eMarketer estimates that penetration will even exceed 60% by 2023. One compelling explanation we’ve heard is that China hadn’t built out all the same brick-and-mortar infrastructure, which made it easier to skip that step to reach the growing middle class. There are much fewer sunk costs in retail store fronts, which means companies like JD.com can spend investment dollars on high-tech fulfillment centers instead (link). China also has the population density to reduce shipping times and delivery costs, which certainly helps with consumer adoption. If the rest of the world eventually caught up to the China of 2019, then non-Chinese e-commerce volumes could easily triple, at a minimum. What’s really interesting is the BAML estimates which show e-commerce penetration in the United States jumped to almost 30% in May of this year. This is obviously a direct result of the global pandemic, and much of the increase is likely transient for now, but it does highlight latent e-commerce potential.
Even if we ignore the global pandemic, it’s clear that e-commerce penetration in the United States has been accelerating for decades (Exhibit I). At this rate, the US would reach 40% penetration in fifteen years. And that ignores the potential change in behavior brought about by the global pandemic, which could accelerate this trend. We have no idea what the specific equilibrium level of e-commerce penetration will be, nor do we have any real idea when exactly that equilibrium will be reached, but we think it’s quite clear that this equilibrium is orders of magnitude higher than current penetration, and will probably take decades to get there. This trend will be a massive persistent tailwind for e-commerce platforms like Shopify.
Shopify’s GMV should increase as subscribers grow, but it should also increase as e-commerce penetration per subscriber increases. We’ve already observed this over the last five years, with a subscriber CAGR of ~50%, but GMV CAGR of nearly 75%. Part of that excess GMV growth is the result of more Shopify Plus customers (larger GMV), but part is certainly deeper e-commerce penetration with existing SMBs. If the industry-wide global subscriber base grows by 5.0x, we estimate that global e-commerce GMV through platform businesses might grow by somewhere closer to 15x with the added benefit of growing penetration.
Unlike the market for subscribers, the market for GMV should see marketplaces and DTC platforms compete head-to-head for share in lots of categories. In the United States, Amazon has incredible e-commerce GMV share – somewhere around 40%. If we include other companies like eBay and Wayfair, that figure is closer to 50% for quintessential e-commerce marketplaces. Then there are companies like Walmart and Costco, which have massive e-commerce businesses, and in many respects are just alternative marketplaces. If we include those businesses, marketplace share is probably in the 60% range. These are all businesses that will handle their own payments, fulfillment, and shipping. Despite a massive head start, we expect that the DTC model will gain share over time, because of its relative infancy and the high appeal to merchants. Even companies like Facebook and Google are likely to feed the DTC model by directing shoppers to merchant websites. Ultimately, we expect all but the largest of enterprises to utilize a third-party e-commerce platform, and to utilize both the DTC channel and marketplaces.
It’s also worth reiterating that Shopify Plus merchants make up a minuscule portion of total subscribers, but the majority of GMV on the platform. A handful of large enterprises can genuinely move the needle. For example, Staples Canada had traditionally managed a custom-built e-commerce platform in-house, but found that they were “running behind from a table stakes perspective”, and decided to outsource the function to Shopify. It’s a great case study (link), but the big takeaway is that one enterprise likely added US$300-400 mln to GMV, which was roughly 1% of Shopify’s 2018 total. There are plenty of large enterprises that still run clunky internally built e-commerce platforms, and we think the vast majority of these enterprises will end up outsourcing that function to third parties like Shopify. This outsourcing shift is yet another tailwind to support GMV growth for the e-commerce platform industry.
Payments
The bulk of Shopify’s 2019 Merchant Solutions revenue was generated by payment services. The company does this through Shopify Payments, which is powered by Stripe. Shopify collects a payment fee from merchants that covers the Stripe technology cost, interchange fee, card issuer fees, and some modest profit for acting as the merchant acquirer. Companies that own the merchant relationship tend to be in a great position to act as merchant acquirers, and in Shopify’s case we estimate that roughly three quarters of merchants have enabled Shopify Payments, with more than 40% of all 2019 GMV being processed by Shopify. Payment penetration has increased consistently since 2013, and we expect that to continue. With growing GMV and increasing penetration, we expect payment services revenue to grow meaningfully over the next decade.
Within Shopify Payments, the company offers an accelerated checkout feature for consumers called Shop Pay. It’s basically the same thing as Apple Pay. Shopify conducted an internal study of a few thousand merchants and found that cart conversion rates increased by 1.7x for merchants that have enabled Shop Pay. We suspect penetration of Shop Pay is fairly low today, but given the obvious benefit to merchants, it’s likely that an increasing proportion of merchants enable this feature. This should help drive more GMV through Shopify Payments.
Shopify is also working hard to be a full omni-channel payment processor, and has provided physical POS software and third-party hardware for years. More recently they’ve launched their own payment hardware for brick-and-mortar stores. Many of the POS services are free, but the new Shopify POS Pro service (includes hardware and software) will start to be monetized later this year on a subscription basis. As a Shopify merchant, it’s valuable to use a single company as the one-stop-shop for multiple business needs, which leads us to believe that Shopify will likely roll out their hardware to a large percentage of total merchants.
Shipping, Fulfillment, and Storage (SFN)
The company has also moved into the shipping and fulfillment business, which should create a ton of value for both Shopify and their merchants. Shopify Shipping was launched in 2015, and allows merchants doing their own shipping and fulfillment to leverage Shopify’s bargaining power with shipping partners to receive reduced rates. Merchants that have adopted Shopify Shipping on the Shopify plan in the United States can reduce shipping fees by as much as 72% (according to the company). Nearly 50% of Shopify’s merchants had adopted the Shipping service as of 1Q20. Over the last 12 months, the company also announced and rolled out the Shopify Fulfillment Network (SFN), which gives merchants the option to outsource fulfillment to Shopify and their partners. The idea here is similar to Shipping: leverage Shopify’s bargaining power, scale, and analytics to reduce the unit cost of fulfillment and the time-in-transit for delivery. To understand the value to merchants, it’s worth going through a theoretical example.
Fulfillment, and Storage (SFN) - Value Proposition
Jane’s Apparel is a fictional retailer with $3.0 mln in sales, which is typical for a U.S. business with 15-20 employees. Jane sells 100 items that come in 6 sizes and a single color, so she has 600 SKUs. Today she manages her own fulfillment out of a leased space. It typically takes consumers 2-6 weeks to receive delivery of online orders, and the average shipping cost is $35/order (we just ordered a bike jersey from a Shopify merchant, and this was the exact experience). Jane can see that many customers abandon their cart when they get to checkout (we almost did), and suspects it’s because of the high shipping fees and long delivery times. Jane is evaluating whether or not to use the Shopify Fulfillment Network to improve conversion for her online business.
To achieve next-day delivery across most of the United States, she needs to keep inventory in 8 different fulfillment centers, and wants at least two units of every SKU. In that case, she needs to stock 9,600 items (600 * 2 * 8). Normally, Jane would keep 4 units of every SKU at her leased warehouse, which worked out to 2,400 items, so her inventory needs increase by 7,200 units. The cost to produce each unit is $20, so total inventory value goes up by $144k (7,200 * $20). She uses Shopify Capital to fund the NWC increase at 8%/year, so her cost of carry is $11,520 ($144k * 8%). She also needs to pay for storage in the fulfillment centers, and the annual storage cost is US$13/square foot. She can fit fifteen items in a square foot, so her annual storage expense is $8,320 (9,600 * $13 / 15). Her total incremental expenses are roughly $20k/year, and at a 10% margin, that means she needs to achieve at least $200k in incremental sales to offset the cost of SFN. This works out to about 7% of her $3.0 mln in annual sales. Is it worth it?
Average shopping cart abandonment rates in the United States seem to range between 70% and 80%, which is crazy high (link). For every five customers that add an item to their cart, only one actually purchases the item! A lot of effort has gone into understanding why cart abandonment is so high, and some of the most popular reasons include shipping costs that were too high, shipping information that was unclear, or delivery times that were too slow. The Shopify Fulfillment Network should simultaneously reduce cost and delivery time, which should obviously improve abandonment rates. If SFN helped a merchant reduce abandonment rates from 75% to 65%, that would result in a 40% increase in e-commerce revenue. If 25% of the merchant’s sales were through the e-commerce channel, that would work out to a 10% (40% * 25%) increase in total revenue. In Jane’s case, this revenue lift from better abandonment rates is already enough to offset the cost of SFN. We’d also need to factor in a revenue increase from higher repeat purchases, and cost savings from outsourcing fulfillment and storage (employee and rent costs). There are plenty of examples where SFN might not make sense for a merchant, but we think that more than 50% of Shopify’s GMV could ultimately benefit from utilizing the network.
Fulfillment, and Storage (SFN) - Partner Network vs. In-House
In our view, there are two ways that SFN could unfold for Shopify. At one extreme, Shopify could rely almost exclusively on a partner network, and not own any of the fulfillment value chain themselves. On the other extreme, Shopify could build out all of their own fulfillment centers (like Amazon).
If the company relied on a partner network, then Shopify would collect a small fee for acting as the network brain and connecting millions of merchants with thousands of partners. It’s also likely that Shopify would lease out 6RS robots to their partners, which would drive down the unit cost of fulfillment for merchants. In that scenario, we estimate that Shopify could earn a fee that might work out to 50 bps for every dollar processed. At the very beginning, we estimate that gross margins wouldn’t be too different from existing Merchant Solutions, but we’d ultimately expect gross margins to be extremely high given the negligible marginal cost of bringing on new partners. In 2019, if Shopify had earned a 50 bp take-rate on GMV through SFN, captured 50% of total GMV, and earned a 70% gross margin, then Merchant Solution gross profit would likely have been 30-35% higher.
On the other end of the spectrum, the company could build out their own network. Amazon currently operates their own fulfillment network, and charges a ~13% take rate for all third-party GMV utilizing the service. The cost structure is obviously quite different than an outsourced model, and our analysis suggests that gross margins are probably in the 25-30% range. If we use this as a proxy for Shopify, and once again assume 50% of total GMV is captured by SFN, then 2019 segment gross profit would have been 300% higher (an additional $1.0 bln). In this scenario, Shopify would have to spend considerable capex to build out facilities. We estimate that $1.0 bln of e-commerce GMV requires around 750k square feet of fulfillment capacity, and that the cost to build/acquire that capacity is around $65-70/square foot. That works out to about $3.0 bln in spending to accommodate 50% of 2019 GMV. Once Shopify got to scale, we estimate that incremental SG&A/Marketing as a percentage of processed GMV would be quite small, which would mean that the payback period on total fulfillment capex might only be 4-5 years, with unlevered IRRs in the 20-25% range. If you’re reading this and think any of our assumptions are out to lunch, we’d love for you to point it out.
For now, Shopify intends to simultaneously rely on partners and build only a handful of their own facilities, which makes sense to us as a first step. It took Amazon more than two decades to build out their own network, and there are lots of hurdles to doing this quickly – capital included. We note that Amazon launched third-party fulfilment services nearly fifteen years ago, and is servicing 70-75% of third-party GMV today. Regardless of how SFN unfolds, it’s clearly going to drive meaningful new revenue growth for Shopify, and is yet another service that should make merchants increasingly sticky.
Merchant Bank
Shopify has two additional offerings that look and feel like a merchant bank when combined. Capital was launched in 2016, and helps fund working capital balances. Shopify either lends the merchant capital outright, or purchases future receivables at a discount. In our view, Shopify has such good visibility on merchant KPIs (both micro and macro level) that they can likely underwrite these often-securitized loans much more effectively than traditional lenders. Shopify can also turn around an application in as little as a day – which is obviously convenient for merchants. By year-end 2019 the company had about US$150 mln outstanding through Capital, but we suspect 2020 will see a step change increase in loans outstanding as merchants lean on Shopify to get through the current pandemic (already hit US192 mln by the end of 1Q20).
On the other side of Capital is Balance, which was announced at Shopify ReUnite in May, and should be rolled out later in 2020. As we understand it, this is basically a merchant bank account. Shopify will likely find a partner to help provide card services, which will enable merchants to make purchases and receive revenue in the same account. The deposit balances are likely to be a funding source for Shopify Capital. And voila, a merchant bank is formed! It’s not clear how the risk and rewards will ultimately be split between Shopify and partners, but one thing is very clear: this is yet another valuable service to merchants, specifically entrepreneurs starting at ground zero. Under the right conditions, this could end up becoming a massive business.
We pulled working capital data that Aswath Damodaran curated at NYU (link) as a starting point for understanding the opportunity set for Shopify. The average relevant business in his sample had non-cash working capital requirements equal to 13-14% of annual sales. If that was a reasonable proxy for Shopify merchants it would mean that Shopify’s subscribers had about US$8.5 bln of NWC in 2019. If 25% of Shopify’s merchants (by GMV) funded their NWC requirements through Shopify, and the company earned a net interest margin of 3.5%, then 2019 net revenue in Merchant Solutions would have been 8% higher. This would also be a business that scales with GMV, and contributes to multiplicative growth in revenue. In a blue-sky scenario, the merchant bank could probably lend much more than for just working capital requirements.
We note that this type of service would contribute to the “sticky merchant” theme. A sticky merchant is also one that might pay 30% more for the e-commerce platform subscription just to avoid the hassle of switching “banks”.
Other
Shopify continues to roll out new services and features quickly, and we suspect there are many more in the pipeline that could ultimately be monetized. One such example is the Shop application that was rolled out earlier this year, and allows consumers to track orders, follow merchants, and use Shop Pay. In many ways, this looks to us like a roundabout attempt to aggregate demand. In the last month, we’ve made purchases from Costco, Amazon, Lululemon, Twin Six, Allbirds, and Bicycle Booth, and have been able to track all of those orders, from all those different counterparties, through the Shop app. We keep coming back to the app to track orders, and while we’re there we see new products from merchants that we’ve previously ordered from. This probably leads to more repeat purchases, many of which will come through the Shop funnel. There is a scenario where the Shop app becomes a meaningful sales channel to merchants that’s exclusive to Shopify. It’s not totally clear how Shop would monetize the app, but one way would be to charge merchants a fee every time a customer clicks through to their online store, or every time a purchase is made through Shop. On the other hand, perhaps the company does something like Amazon Prime, and charges consumers a fee to cover the cost of same-day or next-day delivery. That might be revenue neutral in isolation, but would probably drive greater GMV through SFN. It’s clearly hard to handicap the range of outcomes (one of which might be that nothing happens), but we think there is some option value in a service like this.
Another interesting future opportunity might be an advertising or cross-selling service. None of the Shopify-powered merchant websites we’ve visited have any third-party advertisements. At checkout, there are also no recommendations to purchase items from other retailers. But maybe there could be value in this. For example, we just finished purchasing a pair of running shoes from Allbirds, and perhaps Garmin would want the opportunity to cross-sell the new Fenix 6 watch. If Garmin was willing to pay Allbirds for that opportunity, and Allbirds had no intentions of entering the watch business, a valuable partnership could be created. Shopify sees an extraordinary amount of merchant data, and might see an extraordinary amount of consumer data through the Shop app. In theory, they could be in a great position to facilitate these partnerships for a modest fee. Something like this might be a bit pie-in-the-sky, but it goes to show that once a platform has millions of customers, there are probably lots of incremental services they can roll out at scale.
Competitive Position
In our view, the e-commerce platform space is still going through the land-grab phase. Whatever company has the best product and most effective sales channels should acquire the greatest number of subscribers. The company with the most subscribers can afford to spend the most on R&D and marketing, which helps them acquire even more subscribers, and so on. We suspect that this positive feedback loop will concentrate subscriber market share in the hands of just a few platform companies. For example, Shopify spent more money on R&D in 2019 than BigCommerce or Weebly generated in total revenue. As a result, Shopify was able to roll out a bunch of new free services in 2020 like e-mail, private messaging, POS hardware, and new website features. We’d be shocked if BigCommerce kept pace with Shopify over the next five years from a product development perspective. Shopify also reported that more than 26,000 partners referred merchants to Shopify in the last year, up from just 11,000 in 2016. We estimate that this partner network is directly responsible for 15-25% of Shopify’s subscriber growth, and maintaining the partner network is expensive, because Shopify pays out referral fees. This cost creates a barrier to entry for new platforms. Lastly, like most businesses on the internet, digital marketing is a critical piece of the sales channel. As the biggest e-commerce platform business, Shopify can afford to spend more money with search engines like Google or social media platforms like Instagram. All told, there are clearly some benefits to achieving scale before the competition, and the data from Exhibit D and E support the idea that Shopify has already done this.
We also note that many of Shopify’s competitors don’t offer packages for the full merchant spectrum. As we highlighted earlier, Magento only targets large enterprise, BigCommerce medium enterprise, and WooCommerce small businesses. In our view, this is a huge advantage for Shopify. The enterprise merchant is the most profitable, and there are two ways to acquire enterprise merchants: hire a sales team to go out and onboard existing enterprise; and/or, create a massive SMB funnel, where a few SMBs graduate into the enterprise bucket over time. Magento only has one of these channels, so probably loses share over time. On the flip side, because enterprise customers drive the most GMV, WooCommerce will have a relatively more difficult time amassing the GMV necessary to support something like a fulfillment network. This puts them at a big disadvantage, as we touch on below.
Once a platform achieves critical mass, it can roll out other ancillary services like payments, shipping, fulfillment, lending, etc. Each of these ancillary services requires a lot of something to be successful: a fulfillment network needs to be supported by massive amounts of GMV to sufficiently reduce unit costs and delivery times; payment software and hardware requires lots of R&D and a large subscriber base to monetize effectively; and, lending requires the data and diversification from lots of subscribers in different industries and geographies. In each instance, it only makes economic sense to roll out these ancillary services after some high subscriber threshold is reached, and Shopify is the only platform that is doing all of these things today. If the subscription cost and quality is comparable across the major competitors, the differentiating factor will be ancillary services, so getting there first also matters. In our opinion, the ancillary-services flywheel is much more important than the scale-driven positive feedback loop in the subscription business.
A great quality subscription offering, deep sales channel, and ancillary services, all help capture the customer relationship, and we suspect that customers are incredibly sticky once they join a platform like Shopify. For small merchants, it’s likely the entrepreneur/owners that makes the platform decision. For those customers, it’s not worth spending time contemplating a new platform unless you’re genuinely unhappy with your existing one. Most SMBs that are happy with the platform, and use the unique ancillary services, probably don’t churn unless they go out of business – a $10/month difference in a competitor’s subscription service wouldn’t be enough to move the needle. As an anecdotal aside, we run this blog on Squarespace, and it’s sufficiently cheap that there is almost no reasonable scenario in the foreseeable future that we go through the effort of switching providers. For large customers, the replatforming decision can take a few months, and then the physical implementation of the new platform can take a few more months. Most enterprises won’t want to go through those transition pains unless a competing platform were notably better. Even then, the switching cost (the actual cost, stress on the decision maker, and integration risks) are modestly prohibitive.
Lastly, once a platform has rolled out all these services and owns the customer relationship, the barriers to entry for a new competitor are enormous. Even behemoths like Facebook would have to spend an extraordinary amount of time and capital to compete effectively. For that reason, we’re skeptical that any new platforms will end up as meaningful competitors. We’ve seen some evidence of this already, with Amazon failing with Webstore, eBay dropping the ball on Magento, and Facebook and Google both deciding not to compete. In the race to summit Everest, Shopify is at Camp Four, and new entrants are in Kathmandu.
In our view, the nail in the coffin for Shopify competitors could come through the Shop app, but only if Shopify can successfully aggregate demand. This might be a stretch, but business momentum is certainly working in the company’s favor.
In summary, scale matters a lot, both for the subscription business and ancillary services. Shopify has reached that inflection point and the flywheel should continue to help them gain market share of both subscribers and GMV. The company has a sticky customer, and new competitors are meaningfully disadvantaged. This probably won’t be a winner-take-all industry, but it could very well be a winner-take-most one.
Strategy
If it’s not clear already, we think the strategy is pretty simple: expand the subscriber base, increase penetration of existing services to those subscribers, and roll out new services. Shopify clearly aims to be all things to all merchants in the e-commerce space.
To achieve all these goals the company spends a lot of time and capital on marketing and R&D. They spend a considerable and growing amount on digital and offline advertising, and lots of effort expanding the partner network that refers merchants to Shopify. They have actively expanded their business partnerships to include companies like Facebook and Sezzle, which gives merchants access to new customer acquisition channels and services. In addition to traditional marketing and partnerships, they are growing the sales force that onboards enterprise clients, where the ROI is extremely high. The company also reinvests a growing amount of capital into R&D to improve the product, make it easier for partners to make the referral choice, and for merchants to join. All of this growth capital shows up as an expense, and we can see that those two expense lines have grown exponentially, effectively consuming all of Shopify’s free cash flow. But what’s more interesting, is that the dollars spent per merchant has also increased over time in both categories (Exhibit J). We suspect that spending per merchant will continue to grow over time, which feeds the subscriber flywheel, and will strengthen Shopify’s competitive advantage.
Shopify has avoided changing subscription prices for a long time, with the exception of a one-time change in Advanced pricing in 2016. The strategy, for now, is clearly to keep prices low and get merchants into the funnel on a Basic plan. As surviving merchants grow on the platform, some will inevitably scale into higher priced plans, to a point that subscription revenue increases with GMV. Even before that happens, GMV growth helps earn new revenue on most Merchant Solutions. The long-term value of that funnel clearly exceeds the one-time benefit of higher subscription pricing in the near term.
As for capital allocation, Shopify has been a net consumer of capital since inception. All of their internally generated cash is reinvested in the business (both expensed and capitalized), and the company has tapped public markets multiple times since the IPO. In aggregate, Shopify has issued more than US$4.0 bln in secondary offerings. The bulk of this new capital was reinvested in the product and marketing, but Shopify also acquired 6 River Systems last year to help with the fulfillment network. We expect that the company could issue new equity again as they ramp up spending on the fulfillment network and other services, but they might not need to if the rollout happens at a measured paced.
Performance
Performance is difficult to measure in the traditional sense. Measures like EPS aren’t useful because much of Shopify’s growth spending is expensed. If the company wasn’t growing at all, we estimate that their 2019 net margin would have been 15-20%. But alas, that’s not the case, so instead we measure performance primarily by looking at subscriber, GMV/merchant, and ARPU growth.
We’ve already touched on subscribers, but will reiterate that Shopify has added more than a million net subscriptions to the platform in the last seven years, which works out to a 60% CAGR. During that time, GMV per subscriber has grown by 21%/year, and ARPU (average revenue per user) has increased at 16%/year. In the last three years, ARPU and GMV growth have been similar. In our view, this track record shows that Shopify has been effective at simultaneously adding new customers and receiving more value from each customer.
Financial Position
Shopify has no debt, no preferred shares, no convertible securities, and a minimal amount of other liabilities. The company has been completely funded with common equity. At the end of 1Q20, Shopify had roughly US$2.4 bln in cash and equivalents, and likely has closer to US$3.5-4.0 bln today given the recent equity issue. Cash burn in previous years has ranged from $500 mln to $1.0 bln, and even if they accelerate spending through 2020, we expect that they have ample liquidity to fund growth for a couple years. We wouldn’t be surprised to see another equity issue, but believe that Shopify is close to the self-funded inflection point, after which they might not ever need external equity capital again.
Management & Governance
The management and governance story pretty much all comes down to our view on Tobi, who is the founder and CEO of Shopify. He’s one of the most thoughtful and unconventional leaders we’ve ever come across. He grew up in Germany and dropped out of high school to pursue a computer programming internship, which is a fairly taboo move. Some years later he followed a girl to Canada, where he started an online snowboarding store, which ultimately led to the creation of Shopify. As the CEO of Shopify, there is a long list of unconventional decisions and programs he’s made or led. He offered a college student an internship on Twitter after learning about his StarCraft accomplishments, and then took to reddit to explain his decision (link – scroll down in comments). He put an incredible amount of thought into how to optimize a cafeteria layout, as explained on Invest Like the Best (link – 22:00), which shows an insane level of thoughtfulness and attention to detail. Under his leadership, Shopify helped pioneer the Dev Degree (link), which is a co-op program originally rolled out at Carleton and York University to directly recruit and train a new funnel of Shopify talent. And the list goes on. When a guy like that sets company culture, it’s bound to attract great people. It’s a hard thing to quantify, but we suspect that Shopify has a people and culture edge thanks to Tobi. Other leaders at the company seem to be cut from the same cloth.
As the founder, Tobi is well versed in the product that Shopify sells and the software problems that the company needs to solve. As someone who was once a small business owner, he also seems to understand Shopify’s customers in a very granular way. This deep understanding of the product and customer has clearly translated into a best-in-class platform, and we don’t think that this journey is over. When Shopify IPO’d, and was just a fringe tech business in Ottawa, Tobi wrote something that we think perfectly captures Shopify culture and vision:
“Today, businesses sell through dozens of different channels: online stores, retail stores, wholesale, at pop-up shops, on social networks, through mobile apps or any number of other ways. Merchants often hack together different applications and technologies in order to address their multi-channel requirements. We’re now showing them that they don’t have to; that their complex setup can be reduced to a single, simple platform. By the time we’re done, we think Shopify will have established the ‘new normal’.
I want Shopify to be a company that sees the next century. To get us there we not only have to correctly predict future commerce trends and technology, but be the ones that push the entire industry forward. Shopify was initially built in a world where merchants were simply looking for a homepage for their business. By accurately predicting how the commerce world would be changing, and building what our merchants would need next, we taught them to expect so much more from their software.
These underlying aspirations and values drive our mission: make commerce better for everyone. I hope you’ll join us.”
We love what Tobi is doing, and trust him as a leader and capital allocator, which is really important considering the dual-class share structure at Shopify. Class A shares are entitled a single vote, while Class B shares are entitled ten votes (economic rights are the same). Tobi owns 7% of the economic interest in Shopify, but controls 34% of the votes. In aggregate, the directors and officers of the company, which consists of fourteen people, control 52% of all votes and have a 12% economic interest in Shopify. Additionally, Fidelity owns more than 10% of outstanding Class A shares (4% of votes). These parties clearly have a vested interest in Shopify’s success, but if we didn’t trust those leaders this would be a red flag for us.
Forecasts & Valuation
Subscription Solutions
There are three main assumptions that we make to arrive at subscriber growth in our base case. First, growth in the number of global SMBs is effectively nil, going from 47 mln today to only 51 mln at the end of the forecast period. Second, the percentage of SMBs that utilize an e-commerce platform grows from our estimate of just 13% in 2019 to 66% in 2034. And third, Shopify gains share over the next fifteen years, going from our estimate of 20% in 2019 to 30% in 2034. Exhibit L shows what the e-commerce platform landscape looks like today, and what we suspect it might look like in fifteen years.
In our base case, Shopify reaches 10 mln subscriptions at the end of the forecast period, but subscriber growth slows after 2020. We note that the pie is growing so quickly that even though many platforms cede share to Shopify in the base case, each competitor can still realize meaningful subscriber growth in absolute terms (Exhibit M).
On the pricing front, we assume that Shopify Plus subscription fees grow more-or-less in-line with retail sales growth to reflect the link to GMV. For all plans, we assume two 5% price increases in five and ten years, which translates into a fifteen-year fee CAGR of less than 1%. In real terms, the platform would get cheaper for most merchants.
We estimate that more than 90% of total subscribers are under the Basic plan today. As subscriber growth slows, and merchants on the platform mature, we’d expect to see a greater proportion of total merchants migrate to Shopify and Advanced plans. It’s difficult to estimate what the equilibrium split should look like, but our base case assumes that about 80% of subscribers fall into the Basic plan in the terminal year. This shift drives a 1.5% increase in average subscription fees/merchant through the forecast period.
Lastly, we estimate that gross margins for the subscription business remain more-or-less flat through the forecast period. The net result of these assumptions is a 21% CAGR in Subscription Solution gross profits, from US$514 mln in 2019 to more than US$9.0 bln in 2034.
Merchant Solutions
To arrive at our GMV estimate in the base case, we multiply total subscribers by average GMV/subscriber. We assume that GMV per subscriber increases meaningfully in 2020 as lost POS sales move online, but that most of this increase is transient and disappears in 2021. We assume modest growth thereafter, which reflects a combination of A) retail sales growth of 2.5%, and B) increasing e-commerce penetration.
In the base case, we assume global consumer spending falls in 2020, and doesn’t reach 2019 levels until 2022. In the terminal year we assume that e-commerce penetration of retail sales is 40%, and penetration of non-retail PCE is 12.5%. This results in global e-commerce sales growth of ~10%/year through 2034.
The implication of these estimates is that Shopify’s share of total e-commerce grows from roughly 1% today to 6% in the terminal year. We assume that marketplaces, like Amazon, make up about 50% of e-commerce sales; the travel segment remains cornered by large online travel agencies; and that food delivery services, like Uber Eats, capture the bulk of that segment. As a result, Shopify would see GMV growth of more than 20%/year, and exceed US$1.0 trillion by the end of the forecast period. In 2019 dollars, our terminal year GMV estimate is roughly 2.5x what Amazon sold in 2019, and we suspect one major pushback to our thesis will be how ridiculous that sounds. However, if Amazon’s share of global e-commerce remained flat, and they enjoyed industry-average GMV growth, we note that Shopify’s 2034 GMV would still be 10-15% lower than Amazon in that year. Although they are technically different models, we find this comparison to be a helpful gut check.
Shopify’s Payment business should grow slightly faster than GMV as their penetration increases. Shopify Payments processed around 40% of total GMV in 2019, which continues to grow proportionally since the 2013 launch. We assume that 55% of GMV is ultimately processed by Shopify Payments, which explains about a third of the increase in take-rate through our forecast period. The other two thirds can largely be attributed to the roll-out of Shopify POS Pro. We see no compelling reason to assume that gross profit margins change meaningfully.
We expect that the Shopify Fulfillment Network will become the largest contributor to revenue growth. The base case assumes that 55% of GMV ultimately utilizes SFN, and Exhibit R shows how that adoption curve compares to an estimate of Amazon’s fulfillment adoption by third-party sellers. We note that both the largest and the smallest merchants probably won’t need to utilize SFN; not enough volume on one hand, and enough volume to manage fulfillment internally on the other. We expect that Shopify will lean heavily on partners in the early years of SFN, but that they will ultimately build or buy the majority of their own fulfillment centers (75%). In our model, we assume Shopify reports net revenue (only their share of SFN), so as they internalize SFN capacity, take-rate increases to reflect a greater share of value. At the same time, we acknowledge that the gross margin on internally operated fulfillment centers is probably only half as high as outsourced fulfillment, which explains why our Gross Margin rises in early years but falls off meaningfully as the network is internalized.
There is almost nothing capital intensive about Shopify’s business (in the traditional sense) outside of SFN. As a result, the majority of the capex in our base case is tied to internalizing SFN. At 750k square feet required per $1.0 bln of GMV, and $65/sft, we expect that Shopify will have to spend upwards of US$30.0 bln on the fulfillment network to internalize 75%. We think that Shopify has sufficient liquidity and line-of-sight to meaningful operating cash flows to largely self-fund this entire build. Any external capital needs for the network are likely to be met with bank debt.
Outside of payments and SFN, we make a few other assumptions about Merchant Solutions that are much less important but can be found in our model. Notably, we include very little value for the merchant bank, and no value for monetization of new services.
EBIT Margin
It’s difficult to forecast what Shopify’s EBIT margin might look like in fifteen years, so we tried to tackle the problem from a few different angles. The bottom-up approach had us looking at Sales & Marketing, R&D, and SG&A. Directionally, we expect these costs to grow slower than revenue for two reasons: first, Shopify should benefit from operating leverage as revenue scales; and second, a significant portion of Sales & Marketing and R&D expense is actually growth capital, and as growth slows we’d expect to see the growth portion of these expense items fall. From 2013 to 2019, Shopify’s subscriber CAGR was >50%, while combined Sales & Marketing and R&D spending per merchant grew at 7%/year. In our forecast period, we assume that subscriber growth falls to 16%/year, and that combined spending per merchant still grows at 6%/year. We think that Shopify is already spending more per merchant than other competitors, so this would clearly help the company maintain its market leading position. Exhibit T shows how this cost assumption changes as a percentage of revenue, and results in a terminal year EBIT margin in the 20-21% range.
We also once again looked at Amazon as a proxy, and note that Amazon’s 2019 EBIT margin was 5%, with a gross margin of 41%. Amazon includes fulfillment expenses under gross profit, while we include fulfillment expense in gross profit for our Shopify forecast. Adjusting for this, our terminal year gross margin assumption for Shopify is 17% higher than Amazon’s in 2019, which we suspect should translate into a higher comparable EBIT margin. We note that Amazon’s existing EBIT margin includes growth capital in Marketing and Technology & Content. Making these adjustments, it’s fair to assume a comparable EBIT margin for Shopify in the 20-25% range, which is more-or-less in-line with our bottom-up approach.
Lastly, we plotted EBIT margins vs. gross margins for the North American IT Services sector (Exhibit U). We only included companies with trailing Y/Y revenue growth between -20% and 20%. If we plugged Shopify’s terminal gross margin into the linear regression, we’d get an EBIT margin of about 20%.
We’re confident that margins should expand as Shopify grows the subscriber base, improves their competitive position, and monetizes those merchants. That being said, we do recognize that this is a difficult number to forecast, and would love to hear from the 10th (wo)man on how we might be able to think about this better.
Valuation
Our base case suggests that fair value for Shopify is somewhere in the US$950/share range, which is 17% upside to the current share price. We approached this base case exercise by thinking about end-states, and not worrying as much about every detail on the path there. We also decided on a 15-year DCF to help combat some of the biggest pitfalls of a more conventional 5 or 10 year forecast period.
Shopify definitely has the widest range of potential outcomes relative to any of the other businesses we’ve written about so far. That makes a single mid-point estimate much less useful in decision making, and scenario analysis critically important. Our Bear and Bull case suggest fair value of $335/share (-59%) and $2,350/share (+188%) respectively. Notably, to arrive at the Bear case we had to simultaneously assume lower subscriber growth, lower GMV/subscriber, and a lower take-rate on GMV. That scenario assumes that the merchant bank is a flop, SFN is never in-sourced and penetration remains <50%, no new services are monetized, and e-commerce penetration only doubles from here - it’s incredibly punitive. You can find all of our assumptions for the Bear and Bull case in the model, but Exhibit W shows the primary outputs and how they differ from the base case. Our primary takeaway from this exercise was that there are a lot more things that can reasonably go right for Shopify than wrong. Risk is clearly skewed to the upside.
What would the 10th Man say?
We think that the view and forecasts above are fairly balanced, and suspect that the 10th (wo)man is most likely to sit at either extreme: way more pessimistic, or way more optimistic. In a winner-take-most outcome, Shopify could easily be worth 2-3x our base case. On the other hand, if even one or two other competitors can successfully replicate all of Shopify’s ancillary services, or global e-commerce penetration ends up being only 25%, our base case will look like a blue-sky scenario. While we don’t think either of these extremes is that likely, they are definitely viable views.
Alternatively, the 10th (wo)man might say that it’s not worth making a bet: it’s impossible to know what Shopify’s market share might be in 2030, and the range of outcomes is too wide. If it’s hard to know what the market is pricing in for 2030, then it’s going to be hard to have conviction on where we have a differentiated view. When we don’t have that conviction, the strategy is simply to keep the ball in play, and to do that might mean we never bother trying to make a bet on Shopify. The ROI of our time might be better elsewhere.