Toast:黏性 SaaS——[Business Breakdowns,EP.247]
- Sean Barrett of Counter Global披露其在Toast的15%仓位,并称这是“一代人的机会”:按2027年GAAP市盈率18倍计,买入的是一家营收增速可持续超过20%、EPS增速可持续超过30%的复合增长股。 Toast在约120亿美元企业价值下实现约20亿美元经常性毛利,EBITDA利润率约35%,账上现金接近20亿美元;其DCF给出的合理价值为“每股大约50美元或更高”,而近期股价仅22-23美元。
- 他对“SaaSpocalypse”的回答是历史:当下像2014-15年,当时AWS将开源软件带上云平台,把软件估值压到营收的3-4倍,直到品类杀手“不断交出数字……股价到2016年初呈抛物线式上涨”。 在他看来,垂直行业多租户SaaS的品类杀手因为AI反而处于更有利的位置,而不是更糟。
- Toast的TAM已从1个扩展到5个——核心SMB餐饮,加上企业客户(Applebee's)、商超/酒类/加油站、酒店(Marriott)和国际市场(UK、Ireland、Canada、Australia)——利润率也从2022年的-16%升至约35%,正向40%+这条“底线”迈进。 过去3年,Toast一直是公开市场增长公司中的第97百分位;目前覆盖超过16万家门店,占美国餐饮市场约20%,并拿下约一半新开美国餐厅。
- AI是“Toast成立以来发生的最棒的事情”:Toast IQ将对话式AI与执行系统结合起来,包括智能代理库存管理,约50%的客户每周使用;新推出的Toast Grow营销引擎收费500美元/月,已带来约8%的总营收提升——以平均营收130万美元的客户计算,一开始就是约20倍ROI。 与此同时,公司业务规模翻倍,内部研发投入已连续2年基本持平。
- DoorDash的威胁被带到一线验证:一位同事走访了旧金山30-40家同时使用两套系统的餐厅,“我们找不到一家客户说,如果DoorDash免费提供POS,他们就会切换”。 这笔账“并不理性”——只为支付费率节省约50bps,却要承担13-15%的外卖抽成;Toast与Uber Eats合作推出的免费配送模块估计已有约8万家餐厅注册,反过来让DoorDash陷入创新者窘境。
- 行业每年15%的流失率被重新定义为挑战者的优势——每年约10万-12万家餐厅关门,通常随后重开或由新餐厅接替,为Toast提供了在99%留存率行业中根本不会出现的获客机会;同时,典型Toast客户的利润率约15%,高于典型餐厅的10%,因此更能存活、流失也更低。 增长空间在于:到2035年,毛利从20亿美元增至100亿美元,面对不含中国、拥有1500万家门店的全球TAM,即使按“糟糕的倍数”计算也意味着约4倍MOIC,按NASDAQ倍数则接近10倍。
1. 按价值股定价的15%仓位
- Barrett与Toast结缘于COVID时期的2020年:他第一次投资时打开管理层发来的模型,脱口而出:“这不可能是对的。这些数字太好了。对于餐饮业来说,留存率不可能这么高。”对方的回答是:“没有,你没有漏看任何东西。这就是数字。”如今,Toast已占Counter Global仓位的15%。
- 看穿10-K中交换费会计口径的噪音后,Toast当前约120亿美元企业价值、约20亿美元经常性毛利、约35% EBITDA利润率、资本开支极低,毛利增速超过25%;相比之下,2020-21年超高速增长期的EBITDA大幅为负,股权激励费用占营收30%+。Barrett的核心判断是:“明年GAAP市盈率18倍,买入的是一家营收增速可持续超过20%、EPS增速超过30%的复合增长股。”
- Matt Russell早期追问:这部分毛利真的可持续吗?Barrett给出的诚实区分是,它“在性质上具有经常性”。其中2/3是按净费率确认的支付毛利,1/3来自软件;支付业务的波动性高于合同制业务,但客户平均使用7个模块,“如果客户关掉Toast,就没法经营”,而且支付收入会随通胀和GDP增长。
2. SaaSpocalypse与2015年的历史回声
- Barrett并不否认AI带来的颠覆——“我们并非对此视而不见”,编码和DevOps软件确实暴露度很高;但对垂直行业品类杀手,他更愿意参考2014-15年的先例:AWS把开源软件放上平台,软件估值“跌到营收的3-4倍,基本就是今天的水平”,所有人都在问,为什么还要为软件付费。
- 这轮调整约18个月后迎来反转:到2016年初,“品类杀手不断交出数字,不断创新……股价开始呈抛物线式上涨”。他的结论是:“我们迎来了一代人的机会,可以在GAAP净利润倍数上投资Toast这样的公司。”
- 多租户架构正是AI扩大差距、而非抹平差距的原因:Toast可以“写1次代码,然后用新的AI模型把所有更新快速部署到整个客户群”,而本地部署的竞争者可能每隔几周就要派技术员去更新服务器,适逢OpenAI或Claude发布新模型时更是如此——“祝他们好运,这事不会发生。”
3. 商业模型:49bps、每年1万美元,以及无法走捷径的硬件
- 以一笔100美元的餐饮消费计算,交换费约3%,其中约2.50美元支付给银行和卡组织,Toast净收约49bps;相比之下,竞争对手的费率为75-125bps,Square的毛利抽成率约1%,却没有叠加同等的经营系统。Toast处理约2000亿美元交易额,按总量计已是美国最大的商户之一,仍有通过优化成本结构抬高净费率的空间。软件收费为300-500美元/月;算上全部费用,平均客户年付约1万美元,对应130万美元营收。整个业务按用量收费,硬件是获客型亏损产品,餐厅贷款则是规模较小但盈利的业务。
- 每位客户都必须使用Toast硬件,而走捷径的竞争者已经证明了这层护城河:有人开发了iPad应用,但“事实证明,iPad在餐厅里并不好用”——水、热量和摔坏都是问题;对其中一些竞争者而言,最初的捷径最终“反而导致了它们的消亡”。
- 客户偏好的证据包括约50的NPS,Barrett指出25已经“非常好”;95%的客户愿意推荐Toast,Toast占美国市场约20%,而最让团队“真正兴奋”的数字是:公司拿下了约一半新开美国餐厅。历史上,这个行业最终通常由头部1-2家占据40-50%的份额。
4. AI已进入生产环节,而非停留在试验阶段
- Toast IQ是对话式AI加“执行系统”:过去需要店主耗费整个周末的多门店菜单和价格调整,如今可以实时完成;客户还可按需生成定制分析,智能代理库存管理则会在需要时自动从Instacart Business或其他本地供应商订购新鲜食材。约50%的客户每周使用Toast IQ。
- Toast Grow的变现逻辑更直接:它是一套自动化营销引擎,识别出“波士顿一个清淡的冬季周二”后,会自动发起短信促销、网站优惠和Instagram推送,收费500美元/月;相比之下,餐厅每月支付给营销代理商的费用通常为1,000-2,000美元,效果却很有限。早期用户的总营收提升约8%,“一开始就是约20倍ROI”,并有望让SaaS ARPU提升100%。
- 内部经营杠杆同样突出:过去2年“他们交付的产品比我见过的任何时期都多”,但研发投入几乎没有增长,而业务规模已经翻倍。这为销售与营销、一般及行政费用的规模效应之外,又增加了一个新的运营杠杆来源。
5. 竞争格局:传统系统是份额供体,DoorDash是争论焦点
- Barrett将市场拆开看:约一半仍是传统系统,包括NCR Aloha和Oracle MICROS——“没人想把一台发烫的服务器放在厨房后场”;Square在低端市场占5-10%,主要从传统系统和Clover手中赢得份额,而不是从Toast手中夺取。Fiserv旗下Clover约占15%,但“没有创新,客户已经注意到了”,同时还面临Fiserv杠杆率问题。Toast最终胜出的底层原因是,Barrett研究发现,典型Toast客户的利润率约15%,高于典型餐厅的10%,意味着“餐厅老板最终多拿50%的钱”。
- 针对DoorDash的POS试点——Barrett认为这可能是对Toast与Uber Eats合作推出的免费配送模块的回应,估计已有约8万家餐厅注册——Barrett的实地核查结果很明确:切换系统意味着支付费率只节省约50bps,却要承担13-15%的外卖抽成;从消费者和餐厅合计视角看,订单金额最高30%可能变成费用。一位同事走访旧金山30-40家同时启用两套系统的餐厅,“找不到一家客户说自己会切换”。他的保留意见是,两者可以共存,DoorDash更适合外卖优先的快餐餐厅。
- 行业流失率问题被完全倒转:美国约80万家餐厅中,每年有100,000-120,000家关门并通常重开,行业年流失率为15%,这对挑战者是“巨大的正面因素”——在99%留存率的行业里,Toast根本拿不到这些获客机会。自2019-20年以来,市场一直担心Toast规模扩大后流失率会向行业水平收敛,但数据尚未显示这一点;Barrett将其归因于Toast客户本身更健康所带来的幸存者偏差。
- AI原生创业公司也无法简单复制Toast:一家备受关注的新进入者可能以为,靠AI编码就能快速复制Toast;但“快进到今天,我们在市场上根本看不到它”,公司自己也表示,至少还要2年才能做出类似Toast的产品。即便如此,它还要搭建供应链、硬件体系,以及覆盖美国每个主要城市的线下团队。
6. 毛利5倍增长路径、管理层与可能的破局点
- TAM扩张依赖城市级飞轮:Toast在单个城市的份额从2%升至4%、6%,到10%时,“市场会决定统一采用Toast”;如今份额达到25-30%的城市,新增份额速度反而快于小城市。再叠加企业客户(Applebee's)、酒店业务(Marriott)、商超/酒类/加油站,以及管理层称增速快于美国早期阶段的国际市场,面对不含中国、拥有1500万家门店的全球TAM,Toast有望在2035年将毛利从20亿美元提升至100亿美元,而“我们认为这其实只是一个阶段性终点”。
- Toast的护城河不止是软件:它是任务关键型操作系统,拥有网络效应、行业标准化、数据优势和领域专业能力。Barrett还强调了专用硬件、供应链,以及覆盖美国主要城市的线下团队,这些优势都很难复制。
- 3位联合创始人是在MIT攻读计算机科学学位期间相识的,Aman Narang和Elena Gomez是核心管理层成员。Barrett借鉴Munger的测试框架,从诚信、雄心和创新3个维度考察团队;他回忆2020年Toast只有20,000-30,000家餐厅时,创始人已经在谈“如何走到200,000家餐厅”,当时所有人都觉得他们疯了。公司可能在今年或明年初达到20万家。“这是AI时代你想要的团队。”
- 估值有3种算法:按2027年GAAP盈利计算,市盈率为18倍,扣除约20亿美元现金后实际更低;10年MOIC按“糟糕的倍数”计算约4倍,按NASDAQ倍数则约10倍,2035年GAAP净利润将超过30亿美元,且整个市值都将由现金创造;DCF合理价值为“每股大约50美元或更高”,当前股价为22-23美元。
- 风险判断保持克制:宏观风险——不过在2008-09年,餐饮业也只下滑了低个位数,“人总要吃饭”;流失率收敛风险,他会持续盯住;ARPU对价格的敏感度相对没那么令人担忧,因为增长来自能为客户带来ROI的产品,而不是单纯提价;竞争风险,则是“我们始终压着最近的竞争对手”。Barrett从Toast身上得到的最后一课是韧性,以及“保持饥饿”。
完整逐字稿
This is Matt Russell, and today we are breaking down Toast. My guest is Shawn Barrett, founder and CIO at Counter Global. Counter Global manages a concentrated portfolio of businesses in developed markets, and you may recall Sean from an episode last year where we spoke about a name in the alternative asset manager space, EQT.
Today, we are here to cover a business in a completely different industry, and that is Toast. It's a wide-ranging discussion. I think a lot of people are wondering what software names are particularly interesting at this moment in time, and Sean gets into that and much, much more. Please enjoy this episode.
All right, Shawn, it is great to have you back. Today, we are talking about Toast, which is a name that we previously covered, and we do like to revisit names when there are noteworthy things going on. Maybe the story has changed a little bit, and I think we'll get into some of that here. But maybe we could just start off with a simple introduction to Toast for those who aren't familiar, and your own history with this business and what brought you to it—some of those dynamics just to set the stage.
Matt, great to see you. Thanks so much for having me back. Very excited to talk about Toast with you. It's just a phenomenal business. It's a 15% position for us at Counter Global, so it's also a high-conviction name.
For those in the audience who don't know about it, Toast is really the category killer for F&B point-of-sale and software. They're the operating system for their restaurant customers: super mission-critical, super innovative, category killer. I've had a long history with Toast. I first invested in the business back in 2020 during COVID, and I remember the management team sent me the model. I opened it up, looked at it, and said, “This can't be right. These numbers are too good. Retention is too high for the restaurant industry. I think I'm missing something here.”
I called them and said that, and their response was, “Nope, you're not missing anything. These are the numbers.” So, look, it was a great business back then. I think it's an even better business now. Super excited to break it down with you today.
Yeah, I think you got into some of the words that make it thematically very interesting, which we'll get into. But maybe fast-forward to today and set the stage in terms of the financials of the business. Just a snapshot or overview of where they are today. Any comparison to where they were in 2020 and how much the dynamics have changed? I'm curious if you could share that as well.
Oh my gosh, yeah. The business has changed a ton in the last 5 or 6 years. Today, it's about a $12 billion enterprise value. There is some accounting noise in there, so when you open the 10-K, you'll see that they have to account for interchange revenue as their own. But if you parse through that noise, the business does about $2 billion of recurring gross profit with about 35% EBITDA margins, minimal CapEx, and really high customer retention, as we mentioned, on both a gross and net basis.
It's night and day from 2020. In 2020 and 2021, they were in the hypergrowth stage. Even as a public company in the early days, EBITDA margins were substantially negative. Stock-based compensation was really high, at 30%-plus of revenue.
Today, you have a company that is still growing very, very quickly: 25%-plus gross profit growth. We think that can continue for a bunch of years, but it has high quality of earnings. We're looking at 18 times next year's GAAP P/E right now. We'll get into the valuation at some point, but 18 times next year's GAAP P/E for a durable 20%-plus revenue compounder and 30%-plus EPS compounder. So it's a phenomenal business, and it's remarkably cheap at this point.
On the $2 billion in recurring gross profit, I would have assumed this is a transaction-based business. When you mention “recurring,” is that just a floor level, or is it truly recurring in nature with some type of contractual basis?
I think that's a good thing to touch on. When the company describes its gross profit as recurring, I think it's recurring in nature. Something like two-thirds of the gross profit is going to be from payments, where they get a net take rate. That's how they monetize.
Even though the customers, on average, are using 7 modules, they monetize through payments gross profit. That's about two-thirds of the business, and software gross profit is about a third of the business. It is recurring in nature in that if the customers turn off Toast, they can't run their business. The retention rates are tremendously high.
Payments are a funny thing as far as being recurring in nature. They're definitely more volatile than a contract-based business, but they also grow with inflation, they grow with GDP, and they're really aligned with the customer, where their revenue becomes your revenue as well.
There's a lot of nuance to these phrases, words, and semantics, but it is helpful to understand exactly where that's coming from. In terms of revisiting it today, hearing 18 times next year's GAAP earnings is surprising when you reference those growth rates. But what else would you mention in terms of where we are today and why it's worth bringing up and talking about?
I think the business has really evolved in a way that makes it more attractive today than it was a few years ago. From a valuation perspective, there's been a pretty severe debate in public markets around the SaaSpocalypse, or the “SaaS is dead,” narrative.
While we think there is a lot of disruption out there in the software space, particularly around coding and DevOps in some cases, there are also a handful of category killers, vertical-market winners, and infrastructure software companies that are thriving and in a better spot now because of AI than they were a few years ago.
It's important to touch on the SaaSpocalypse debate because we aren't blind to this. There's a real transformation in the market with AI. But when it comes to vertical-market, multitenant SaaS category killers, this actually rhymes with the movie that we lived through in 2015. I think we should talk about it for a minute.
If you go back to 2014 or 2015, open source had been around for a long time, but AWS started putting open-source software on its platform, and it became available to the masses. Pretty much immediately, public-company software multiples collapsed to 3 or 4 times revenue, basically where they are today.
Everyone started asking the question, “Why would anyone pay for software when we have these free, open-source models on AWS?” Sounds pretty familiar. That was a tough time to be a software investor in public markets.
It took about 18 months, and when you fast-forward to early 2016, the category killers kept putting up numbers. They kept innovating. Some of them were using open source internally—most of them were—and the stocks went parabolic.
There are some differences today, no question, but it rhymes with history. At this point, we feel like we have a generational opportunity to invest in something like Toast: a 25%-plus compounder and category killer at a GAAP net-income multiple.
When you talk about the transformation of the business, particularly when you mention the change from 2020 to today, what stands out the most in terms of that transformation and what's happened?
This business has really evolved, and at this point it's firing on all cylinders. I think the biggest evolution that you're seeing today is the result of a ton of hard work that happened 3 to 5 years ago.
So if you go back to 2020, it was an unprofitable business with a single TAM, really just focused on the core SMB restaurant customer in the US. You fast-forward to today, and it's a very profitable business, with 35% EBITDA margins and growing, and 5 TAMs that the company has unlocked through a ton of innovation.
It started with SMB restaurants. They innovated starting 5 or 6 years ago, building products for the enterprise and building products for grocery stores. Now they're selling into liquor stores. They've started quietly selling into gas stations. That's a new opportunity. They're selling into hotels, food and beverage and retail in hotels.
Then they've built an international team and an international product that's already live in the UK, Ireland, Australia, and Canada, and they'll keep adding markets from there. What you had 5 years ago was a single TAM, highly unprofitable, hypergrowth business. Fast-forward, and this business has been a 97th-percentile grower for the last 3 years in public markets, and that's against a lot of hypergrowth semiconductor names.
A 97th-percentile grower while also taking margins from negative to substantially positive, while also innovating like crazy and expanding into a bunch of new markets. It's incredibly rare to find a 10X opportunity in public markets. We think we have one here just based on the current product offering.
In terms of that offering and what they're doing for the customer, can you give that overview now in terms of everything that they're offering to their customer base? What's actually happening as they're selling in? What's on the other side of that offering?
Toast is really the full operating system for their businesses. Think of it like Shopify for restaurants. It's the point of sale, digital ordering, payroll, automated inventory management, real-time reporting across multiple locations, and then they have complex hardware. We should get into the supply chain, but they've built a really great hardware business as well.
Because it's a modern, multitenant SaaS solution, they can code once and then deploy all the updates, with new AI models and new technologies, rapidly to the whole customer base. They've got a huge data advantage that's really important as they go build new products, and that's already starting to show up in their AI offerings.
Most people underestimate just how complex the customer workflow is at a restaurant. But most importantly, customers really love Toast. We do a ton of custom survey work on all our companies. Some notable takeaways as to why customers love Toast so much: Toast has a net promoter score of roughly 50, with 95% of respondents indicating they would recommend Toast. Just for reference, a net promoter score of 25 is considered really good. Thirty to 50 is considered exceptional.
They've got elite customer satisfaction metrics. On average, customers are using Toast for 7 modules, so this isn't just a point-of-sale system like you might see with some of the competitors. It's a multiproduct, mission-critical operating platform for its customers.
Toast is sitting at about a 20% share of the US restaurant market, with more than 160,000 locations, but they're actually winning roughly half of all new restaurant openings in the US. That's what really gets us excited. It's the definition of product-led growth that's leading to very rapid market share expansion, and the network effects in this industry have proven to be really strong over time.
Usually, the top 1 or 2 players in the restaurant industry end up with 40% or 50% share. It's product-led, and they're gaining a lot of share on the back of the innovations that they made over the last 5 or 10 years.
Anybody that's spent time working in a restaurant as a high school or college kid, being front of house, or even just seeing how the point-of-sale systems work there, understands how much flows through when you step back and think about perishables and inventory management. It runs your business. It's notable to think about that.
You helped explain a little bit about the revenue model—recurring and reoccurring. With all of those different offerings coming into play now, what does the revenue model look like, just in terms of what's transaction-based versus paying for the software? How do they go about that?
The revenue model is really a mix of a bunch of the different offerings they have. From a high level, you have payments gross profit, SaaS or software ARPU, hardware, and then a small portion of revenue from lending to restaurants to help them grow. That's a nice, profitable business.
The whole business is consumption-based, so it aligns well with the customers. It aligns well with where the world's moving with AI. But let's break down those revenue line items for a second.
If you start with the payments piece and you look at, let's say, a dinner order that's $100, something like 3% of that will be interchange. $2.50 will go to the banks and the networks and other costs as part of the interchange, and then about $0.50, or 49 basis points, goes to Toast as a net take rate.
That 49 basis points has been moving up over time, as it's reasonably under-monetized versus what you see in the rest of the space. Most competitors charge 75 basis points to 125 basis points from a net gross profit take-rate perspective.
Toast also has some opportunities over time to optimize costs as they get bigger. They've got about $200 billion of volume on their platform now. That makes them, in aggregate, one of the biggest merchants in the US, so they should be able to optimize over time and increase that net take rate.
On the net take rate, that $0.50 versus $0.75—that delta there—who is that negotiated with?
It's effectively what they make from the customer, from the restaurant, after all is said and done. So if an interchange fee is 3% and 2.5% goes to everyone else, they can take the 50 basis points, or the $0.50, at the end.
But it's not a big negotiating point with restaurants. They tend to see it as reasonably priced, certainly against what they see in the market.
The alternative would be charging 3.25%—is that where it's netting out for the restaurant? Is that 75 basis points showing up there?
It could, and there's also a mix of card-present and card-not-present. The mix of credit and debit actually matters when you get into the nitty-gritty.
But if you look at Square, for example, the last time I looked, Square was monetizing at something like a 1% gross-profit net take rate across their business. Square has a nice business at the low end of the market for smaller merchants, but they're not offering a robust, Toast-like operating system on top of it.
That's a pretty pure payments comp that you can look at, and they're monetizing at about a 1% gross-profit take rate.
Understood. I'll let you continue with the software side of things.
On the software side, we talked about how robust the operating system is. Toast charges something like $300 to $500 a month to its customers to use the software suite as well. Depending on the customer, they might use different modules. They tend to bucket it into good, better, best-type packaging.
At the end of the day, when you combine those 2 things together, the customer is paying something like $10,000 a year to Toast on a total base of $1.3 million of revenue for the average customer.
Hardware is a loss leader, so they do charge a nominal fee for their hardware when a customer starts up the business. There are kitchen display systems that connect the front and the back of house. There's Toast Go hardware. It has to be pretty advanced so it can deal with water and liquids dropping and make sure it doesn't break. It's actually a complex engineering feat to build this kind of hardware.
Then they make a little bit of money from lending to their customers to help them grow. From an investment standpoint, it's a very easy business to model and predict, which is always nice from my seat.
So you look at the 5 things that matter here: location count, payments, SaaS ARPU, margins, and multiple. All of those things historically have been pretty relatively predictable. We like it from a predictability and modeling perspective as well.
On your hardware point, does every customer have to have Toast hardware? Is there anything that can run purely on their software?
Every Toast customer has to have the hardware. Toast, when you walk into a restaurant, will usually have some sort of payments dongle at the front desk. You might see some sort of bigger screen that the customer or the restaurant owner is using. Then you have a kitchen display, and you have Toast Go handhelds for the waitstaff.
Toast went down that road of building specialized hardware many, many years ago. It's funny because a lot of the competitors that came out over the years tried to leapfrog them and take the easy road, which was building an app and then asking their customers to just download the app on an iPad.
It turns out iPads don't work very well in restaurants. They break all the time. They don't do well with water, liquid, and heat. What was a shortcut at the beginning for some of those competitors actually led to their demise or made it so that they couldn't gain share.
In the end, the customers came back and said, "This hardware from Toast is actually really powerful. It's purpose-built for the restaurant, and I wouldn't want to use anything else."
Interesting, as it relates to the SaaSpocalypse as well, and how hardware makes a pretty big difference with some of these industries, particularly those that can use it for their updates and whatnot.
So interesting to hear there.
I totally agree with that. It's funny: it's not like a total halo business—hard assets, low obsolescence—but there are aspects of it. You have substantial hardware involved, physical presence, and feet on the street. You're in the four walls of a brick-and-mortar restaurant. So, to your point, I actually think it's an important part of the story.
On the below-the-line profitability dynamics, how has that evolved over time? What have been the big drivers in terms of the changing profitability profile?
I think this is probably a part of the story that deserves more time. At Counter, we talk a lot about product and profits: companies that can lead with innovation and product-led growth, but also do it with really high profitability. There aren't that many companies in the world that can sustain really high growth with great profitability at the same time.
Toast is definitely one of those companies. We talked about 97th-percentile gross profit growth in public markets. That's pretty exceptional. Management says they can maintain 20%-plus growth for many years ahead. They just said that yesterday at a conference again.
At the same time, margins have gone from negative 16% in 2022 to roughly 35% today, going to 40%-plus, which management has said is a line in the sand, and then they'll continue from there. There's a ton of operating leverage in this business. They've seen operating leverage certainly in sales and marketing, where that number has come down a lot as a percentage of revenue. G&A, naturally, as a company scales, has come down quite a bit as a percentage of revenue.
Then R&D has been a really interesting new lever for them as they've incorporated AI into the business internally. They've shipped more product in the last 2 years than I've ever seen them ship. We'll get into their AI offering—it's super advanced and very impressive. In the last 2 years, R&D expense in dollar terms has barely grown. So you've had a business that's doubled in size, and R&D has been relatively flat.
They're starting to grow R&D again, and they'll continue investing. This is an innovative team that thinks super long-term. They're not going to sacrifice their long-term opportunity, but there's been a lot of operating leverage in the business, and we just see that continuing as the business scales.
Maybe you can get into that AI offering and what they're shipping. How does that show up? It's interesting to hear—we hear a lot about experimentation, and I think we're all doing a lot of experimentation. It's another thing to ship things that your customers are using. So how does it show up in the business, and what does it look like for customers?
AI is probably the best thing to happen to Toast since its founding. The product has always had advantages versus the incumbents and the competitors, but you fast-forward to today, and the product has substantial data advantages. They have 160,000 restaurants on the platform, so they see local data everywhere.
They can see that because it's a multitenant SaaS cloud platform. With multitenant SaaS, you can see all the data in real time, and you can give product updates in real time. Then the gap against your competitors gets much wider.
If you look at an on-premise offering, OpenAI is coming out with a new model every week. Claude is coming out with a new model every week or two. If you're an on-premise platform and you have to send a technician out to a restaurant to update the server every couple of weeks, good luck. It's not going to happen, and it doesn't make any sense. It's not rational from a financial perspective.
So AI is widening the gap for Toast versus its competitors. To give you a couple of examples of where they're innovating and what they've done in AI, the first thing they came out with was something called Toast IQ, which is their AI offering. Customers love it. It's effectively a conversational AI offering combined with a system of action.
You can talk to Toast IQ. You can ask it questions, and you can make menu changes across the board. That sounds simple, but if you have multiple locations and you want to change prices or menu items while also connecting to the marketplaces, that used to take an owner all weekend. Now you can just talk to Toast IQ and make the changes in real time.
You can have real-time analytics and custom analytics. Let's say you have multiple locations and you want to say, "Hey, Toast IQ, can you tell me what this location is doing year over year with these menu items?" You can break it down in a bunch of ways.
It also has real-time inventory management. If the restaurant's running out of a certain kind of produce at night, the system will automatically order fresh produce from Instacart Business or other local vendors. It's agentic in nature—it's an advanced AI model—and the customers absolutely love it.
We've heard that 50% of customers are actively using Toast IQ on a weekly basis. The new offering that just came out that's super exciting is called Toast Grow, and it's within the AI offering. It's an automated marketing engine for restaurants.
Restaurants spend a lot of money on marketing. Generally, a restaurant will spend $1,000 to $2,000 a month with a marketing agency for Instagram and local marketing, and they don't see a lot of uplift. They still have nights of the week that are going to be empty or quieter by nature of local behaviors.
What Toast Grow does is look ahead, take data from the past, and take data from restaurants around you. It might say, "Okay, this next winter Tuesday in Boston has historically been really, really quiet. Let's go out and do a local promotion with SMS texting to people who've been here before. Let's put up deals on the website. Let's do an Instagram promotion."
It does it all automatically in real time, for about $500 a month. Historically, just over the last month or two as people have started using this, they've seen an 8% uplift in total revenue. Remember, we talked about a restaurant on Toast doing about $1.3 million in revenue on average.
So this product costs $500 a month as a SaaS module. It's a nice uplift and could be a 100% uplift to SaaS ARPU. It gives you about a 20x ROI right out of the gate. That's the kind of stuff they're doing in AI.
We talked about some of the internal leverage they're getting with R&D as well, but the product roadmap is largely AI-driven and agentic-driven. Management would tell you, as they did at a conference yesterday, that this is going to be the next leg of their revenue opportunity as well.
The 8%—whether you measure it in terms of a 20x ROI or just think about restaurants and the margins they operate at, how razor-thin they can be—is material for that customer base.
On the competitive landscape, I truly failed to appreciate how much Toast was doing for customers, just in terms of how deep into the restaurant's operating activities it could be with inventory management, marketing, and all of these different things. How do you define the competitive landscape? I'm curious how many others are doing quite as much in terms of the breadth of different activities.
The competitive landscape, I think, is a real strength of the story here. Before we jump into specific competitors, the big question is: Why does Toast win? They win because they have a great product, which we talked about; they have happy customers; and, importantly, they have more profitable customers.
A typical restaurant might have 10% margins. It's a low-margin business, as you mentioned. A typical Toast customer, from our work, has something more like 15% margins. So they're taking home 50% more dollars for the restaurant owner. It also means that Toast restaurants survive more often. That leads to higher retention for Toast and happier customers.
The interesting thing about the competitive set is that I would break it into legacy and modern. About half of the market is still on legacy platforms. That's stuff like NCR Aloha and Oracle MICROS. Those are companies with on-premise software and on-premise servers. Generally, as we talked about, that's a bad fit for a kitchen. You don't want a hot server sitting in the back of your kitchen or in a closet near your kitchen.
That's what really opened the door for Toast to be the first successful scaled, cloud-based disruptor. On the more modern side, you've got Square. They have about 5% to 10% of the restaurant market. They do really well at the lower end, or smaller end, of the market, where it's a less advanced offering—really more just a payments dongle.
Now they are innovating, and they're innovating at a more rapid pace than they used to because I think they see the restaurant industry as really attractive. They aren't winning from Toast. They're mostly winning from legacy platforms and, we think, from Clover.
Clover has about a 15% share. That's the subsidiary of Fiserv. It's been publicly noted that Fiserv has been overlevered. They've acquired a lot, the stock has been a mess, and I think they have some real issues to deal with regarding leverage if they want to survive.
As a result, they're not innovating, and customers have noticed. So they've been losing share even on the modern side. One player to watch right now is DoorDash. It's no secret that DoorDash has been piloting POS solutions with its customers, but I think maybe they did that in response to Toast, which very famously, about a year ago, started offering free delivery.
That’s a real innovator’s dilemma for DoorDash all of a sudden. DoorDash has a take rate of close to 15% for its deliveries, but all-in, from a customer and restaurant perspective, it can be 30% of the order that goes to fees. Toast just flipped that whole thing on its head. We think 80,000 customers, or half of the Toast restaurant base, have now signed up for the Toast ordering module, which effectively gives you free delivery in partnership with Uber Eats from the restaurant perspective.
We always try to go out and do our homework on the ground, and so we ask the question: Which restaurants would be likely to switch from Toast to DoorDash if DoorDash were to offer POS for free? There are a few things worth noting. First, the unit economics of switching from Toast to DoorDash are irrational. You’re effectively saying you might save 50 bps on payments, which is not substantial, by going over to DoorDash, where you’re paying a 13% to 15% take rate on deliveries. And that’s a really meaningful difference, that take rate.
You can actually do this, Matt. If you go to your favorite local restaurant, there’s a decent chance now that they have Toast deliveries and DoorDash deliveries enabled. Pull up the delivery on DoorDash, and you can get all the way to the order page. A $40 delivery on DoorDash will usually cost you something like $30 to $35 if you continue to the delivery page on Toast. It’s a really great offering.
Restaurant customers love Toast. Net Promoter Scores of 50 and customer-satisfaction rates of 95-plus percent. Those customer-satisfaction rates are substantially higher than what we see in our surveys on DoorDash. The third thing is that we always put boots on the ground to test our hypothesis. My colleague actually went up to San Francisco the other day, just this week, and walked into 30 or 40 restaurants that have DoorDash and Toast enabled. He bought something to be a good customer, but asked them, “How likely would you be to switch from Toast to DoorDash if it was free?” We couldn’t find a single customer who said they would switch if DoorDash offered free point of sale. We think both these companies can coexist, grow, and be successful. Where I think DoorDash will find more success is in quick-service restaurants that are delivery-first restaurants, versus where Toast plays really well in the full-service neighborhood restaurant.
I can speak from personal experience. The frustrations that have existed with delivery fees from restaurants over the years are quite strong, too, and vocal, and it’s just not something you hear as it relates to any other part of the business. I think that’s out there. I respect the Channel Checks 101, boots-on-the-ground approach that remains alive. That’s incredibly interesting information to get back.
On that whole concept of switching—and whether we call it churn—how much does the industry switch year to year? I’ll almost put aside the fact that the restaurant industry sees a lot of turnover. We have a lot that goes out of business and a lot that comes into business. But do you have any concept of how sticky it is and just general churn?
It’s pretty well documented. The industry churns at something like 15% a year. And so, if you think about 800,000 restaurants in the US, something like 100,000 or 120,000 of them shut down and then generally restart every year. It’s a tough business. What you see is that the business is really tough the first couple of years, and then restaurants that survive for the first 3 to 5 years actually tend to survive for a very, very long time.
The interesting thing from a churn perspective is that it’s actually a huge positive for the challenger, like Toast. There are some industries out there with 99% retention, where there’s no churn. And even if you have a better product, you come in with a hugely advantaged product, customers aren’t going to churn, and you’re going to get 1% of the industry. You’re going to get 1% at-bats every year. You look at Toast as the challenger. Let’s say 100,000 restaurants reopen or open every year. Toast is winning about 50% of those, we think, on a gross basis.
They wouldn’t get all those at-bats if it were a higher-retention industry. It’s a strength of the challenger. I think the big debate, or the big question, is: What happens in 5 years when Toast is 30% to 40% of the overall market? Do they look more like the market, and does their churn go up? That’s been a debate since 2019 or 2020: As they get bigger, surely their churn will go up.
We haven’t seen noticeable changes in churn. And the reason for that, we think, when we talk to customers, is that their businesses are so much more profitable using Toast than they were otherwise that they just become healthier businesses that survive more. So there’s a big survivorship bias in the industry within Toast customers.
From the broader perspective of competitive advantages and moats, we talked a little bit about hardware, software, and logistics. I guess, to the extent that we’ve mentioned a lot of varying things, how would you package that together and just talk about their moat?
There’s been so much negative news on software in the last 6 months. We really wanted to simplify it, and we always come back to what we call the laws of physics here at Counter Global. Charlie Munger used to call them his mental models. But these are rules that have been durable for investing for decades, and they hold true. And so we rely on those superpowers. We rely on those laws of physics.
Toast is a mission-critical operating system. They have strong network effects. The industry is standardizing on this platform. They have huge data advantages. They have domain expertise. Speaking of Charlie Munger and his mental models, one of his mental models that he used to talk about a lot was that in any capitalist system, the specialist wins an outsized share of the economics. He famously quoted that in a lot of his speeches. We believe in these laws of physics, these mental models, and the interesting thing for Toast is that it really checks all the boxes.
The 2 structural moats that I think people don’t talk about enough are hardware and how complicated it is for Toast to deliver this great hardware, and then the distribution advantages they have. Starting with hardware, it took them many years to build purpose-built hardware for the restaurant industry. But it’s not just about the design. It’s about the supply chain, the chips, and making sure you have the right amount of inventory for your share gains. It’s very, very complicated.
And then the distribution is also really complicated. Toast has feet on the street in every major city in America. And those advocates walk into restaurants all day long, talk to owners, and make sure they’re being served properly by Toast. It’s a huge part of the story that’s really hard to replicate. I would just give you an example. A year or 2 ago, a high-profile startup launched. It was supposed to be AI-native, serving the restaurant space. And I think they thought, probably with AI coding and all the advances going on with AI, that they could copy Toast’s product and have something up and running pretty quickly.
Fast-forward to today, we aren’t seeing them at all in the market. Last I heard, they extended their roadmap. They think it’s going to be at least another 2 years before they have something that looks like Toast, and then, at that point, good luck building the supply chain, good luck building the hardware, and then you’ve got to get feet on the street in every major city in America and abroad. I think it’s just way harder to disrupt this space because the moats are so powerful.
We haven’t talked about the management team or just culture within the business. Tell me a little bit about who’s behind it and anything that’s unique about the culture, to the extent there is something unique.
It’s a really phenomenal culture. The restaurant business has had a lot of ups and downs over the years. It’s a culture of people who are really resilient. They put the customer first. So it’s product-led, resilient, and customer-first, and that’s how I would describe their culture.
From the top, it’s a founder-led team. It’s a handful of really, really good people running this business. At Counter, we evaluate our management teams on 3 pillars. We talk about this a lot. It’s integrity, ambition, and innovation.
So when you think about integrity first, and you look at this management team, Aman Narang and Elena Gomez, they’re really great managers. They tell it like it is. They’re super-high-integrity people, and the people around them are super-high-integrity. On ambition, it’s probably one of the more ambitious teams that we see in public markets, and they’ve been this way for many, many years.
In 2020, I remember, I think they had 20,000 or 30,000 restaurants on the platform, and they were talking about their path to 200,000 restaurants and then their path to 400,000 restaurants, and everyone thought they were crazy. The TAM can’t support that. You’ll never innovate to open up new TAMs. It’ll be really hard. Well, they’re probably going to hit 200,000 restaurants this year or early next year.
Now they’re doing $2 billion of gross profit, and they’re talking about their path to $10 billion of gross profit, and I think there’s some skepticism out there. They have the track record of doing it. And so that gets to the third pillar we think about, which is innovation.
This is the team you want in an AI world. It’s 3 co-founders who met getting their computer science degrees at MIT. They really understand the technology. They really understand AI, and I can tell you from spending time with them, they’re more invigorated and more excited than I’ve ever seen about the opportunity because it allows them to accelerate their roadmap, open up new TAMs, and just serve the customer in a way that wasn’t really possible a few years ago.
On that $10 billion gross profit milestone and how you think about TAM, what is the runway here? Where does it come from? Is it U.S.-based? Is it international? Talk a little bit about the path further out than the next 3 to 5 years, maybe.
From a big-picture perspective, the business model is a beautiful compounder. The things that matter are location count, SaaS ARPU, payments ARPU, margins, and multiple. If you double the location count and then also grow your ARPU, you actually have the chance to 4x and then 8x the business over time.
But we care a lot about the building blocks that get you there. That's how we need to underwrite it based on each specific TAM. The beauty of the Toast story is how they've sequenced their growth into new TAMs and opened up new TAMs over time.
We talked about the core SMB as the primary driver of the business many years ago. At this point, probably 150,000 of their 160,000 locations are in that core SMB U.S. market, and they're winning 50% of new restaurant openings. On a local level, I would say the trends are even more impressive.
What you see when you go into the city-level data is that they usually start with a 2% market share, then maybe 4% the next year, and 6% the year after. But when a market hits 10% market share, it becomes what management calls a flywheel market, and that's when the network effects take hold and the market decides it will standardize on Toast.
Then you actually see market shares accelerate faster as they get bigger. We have data on some cities right now that are at 25% to 30% market share, and they're actually adding more market share in those cities than they are in the smaller cities. It's a great playbook.
We think there's a long runway to go just in that core SMB, and certainly management thinks so as well. The interesting thing from there is that they've already innovated and built the product for 4 or 5 new TAMs beyond that, and now they're actually in market. It's all about execution.
They're in market with a grocery store offering for grocery, liquor, and gas stations. That market is primarily served by legacy technology, where Toast has a huge right to win. Customers love the offering. We've talked to a lot of customers, and we think there's a long runway there.
They've started to win in hospitality. You saw a big win with Marriott, so they're winning hotel, retail, and F&B. That's a new offering in the last couple of years. Enterprise is really cooking, no pun intended, where they've won Applebee's and a handful of other really high-profile enterprise customers that would've seemed impossible many years ago.
For people who thought that the TAM for Toast would top out at 400,000 or 500,000 because they'll never get the enterprise, the TAM in the U.S. is actually bigger than the restaurant market. It's enterprise, SMB, plus grocery plus hospitality.
They also quietly built an international business where they're in the market in the U.K., Ireland, Canada, and Australia. Management would tell you that those markets are growing faster than Toast grew in the U.S. when they started here.
I was in London last week seeing companies. You're seeing Toast all over the place, and I wasn't seeking them out. I was just walking into random restaurants, and Toast is all over the place. They're winning there.
Big picture, Toast has 160,000 locations on its platform today. Even if you take out China, they have a 15 million-location TAM globally, most of which is served by legacy technology.
So we think they're just scratching the surface, and they have the opportunity to take themselves from $2 billion of gross profit today to $10 billion in 2035. But we actually think that's just a stopping point. They can keep going from there.
When you think about valuing a business with that type of growth runway and ambition, what's your general approach for valuation for this business?
There are 3 ways that we look at valuation here. All of it is in an effort to get back to intrinsic value. The first way you can look at it is on a multiple of 2027 GAAP earnings, which we talked about. It's trading at 18 times 2027 GAAP earnings.
That would be an even lower multiple if you want to adjust for the almost $2 billion of cash they have on the balance sheet. For a 25% gross profit grower and a 30%-plus EPS compounder, that screens tremendously cheap.
More importantly, we think about intrinsic value and what kind of MOIC we should expect, or multiple on invested capital, over the next 10 years. We have our TAM build, we have our segment build, we know the win rates, and we think it's a highly predictable outcome from a business perspective.
Over the next 10 years, they're going to take gross profit, we think, from $2 billion to over $10 billion. At that point, you'll be left with a business that's doing over $3 billion of GAAP net income, and you will have generated your whole market cap in cash.
So if you get a horrible multiple in 2035, you have something like a 4x MOIC. If you have a NASDAQ market multiple, you probably have something closer to a 10x MOIC.
Then, just on a DCF intrinsic value today, if you want to discount all the cash flows back, fair value today is something like $50 per share or more versus the recent stock price at $22 or $23.
No matter how we slice it, near term or long term, we're really excited to continue to be on this journey with Toast, and we think it's going to be a good outcome for investors.
What are the risks besides the obvious execution? What stands out the most to you?
There are 3 or 4 risks here that we think about, talk about, and debate a lot. One is just macro exposure, and you hit the nail on the head early in the call. How much of that fintech gross profit is really recurring?
I think in this business, in the restaurant business, it is a tough industry in some ways. In other ways, it's also a really great industry. It's a big market with $1 trillion of volume every year. It does grow.
In tough times, we've actually seen the restaurant industry is super resilient. If you go back to 2008 and 2009, I think the restaurant industry as a whole was down low single digits, one of the better-performing industries in the economy, because it turns out that no matter what's going on, people need to eat.
Even in tough times, they want to go out, and they want to eat and socialize with friends or family. The second thing we think about a lot, which we talked about today, is industry churn and how that plays out for Toast over time.
We follow the data, and so far, what we've seen and what we believe we'll continue to see is industry-leading gross retention for Toast versus the competition. But it's certainly something we want to stay on top of.
The third thing we think about is the price sensitivity of the customer base and how much ARPU you can really squeeze out of the customer base over time. If this was a price game and we were betting on a lot of growth to come from pricing, we would be more concerned about this.
What we've seen from the Toast management team, which, again, is super innovative, thoughtful, and customer-first, is that they're coming out with products that can be meaningfully high ARPU for Toast, but also a really good ROI for the customer.
Toast Grow marketing is the perfect example. $500 a month sounds like a lot when you're talking about a customer that probably has $150,000 or $200,000 of profit per year. But then if you put it in the context of raising revenue for your customer by $100,000 and replacing another cost bucket that they have where they're already spending $10,000, it actually becomes a no-brainer, and it's a win-win for Toast and the customer.
And then the last thing—I saved it for last, but it's probably the risk that we stay on top of the closest—is competition. We're always talking to customers. We're always surveying customers. We go put feet on the street and boots on the ground to make sure that Toast is winning in the wild, but it's something that we stay super close to.
This has been timely and fascinating, and Toast has extended way further than I had imagined before researching the name. What stands out to you as a key lesson from this business that you could apply elsewhere?
I've known the Toast team a long time. The restaurant industry is really hard, and Toast has been through a lot over the years with COVID, now the SaaSpocalypse. They always come out on top.
So why is that? I was reflecting on why that is, and what they've reinforced for me, I think, is a lesson of resilience. Work hard, expect the unexpected, put the customer first, learn from your mistakes, and I'll leave you with a pun: stay hungry.
I love it. Well, Shawn, thank you again for coming on, sharing the knowledge, and making it timely. This has been a true pleasure.
Thanks so much, Pat. Awesome seeing you.
Hosts and podcast guests may maintain positions in the securities discussed on this podcast, and actual investment positions taken by podcast guests may vary from the conclusions discussed. There is no consideration given to the specific investment needs, objectives, or tolerances of any of the listeners, and accordingly, this material does not constitute a personal recommendation.
The specific investment examples discussed during this podcast are included for illustrative purposes to indicate the podcast guest’s potential investment processes and strategies, and the types of companies that the podcast guest believes are representative of a particular theme. It should not be assumed that any investment discussed herein has been or will be profitable, or that recommendations made in the future will be profitable or will equal the investment performance of the specific investment examples discussed herein.
Projections represent hypothetical performance and do not reflect the performance achieved or projected by the podcast guests. There is no guarantee that such performance will be achieved, and actual results may vary.