SaaS Metrics Masterclass: Pricing Strategies and Billing Models with Stripe

Thank you very much. Hi everybody. Good to see you today. Thanks for coming. So growth of the kind of subscription e commerce industry has been over 100% year on year for the past five years. According to McKinsey, this wasn't the case. I mean, this is quite new. It wasn't the case 20 or even 10 years ago where the business models of the Internet were more focused on e commerce marketplaces or even advertising. So the question is kind of, first question is, what made SaaS so successful? If you kind of think about that question, thinking about the stakeholders and the decisions in companies of using SaaS products, there's kind of three types. One is customers. Customers love SaaS products and tools because it simply works. You don't need to maintain limit. Errors are very limited and it simply works. The other piece of it is there's a perception that SaaS products are less expensive. Now, whether in actual dollars. It's true or not is not the only question. It's also a perception of risk. Basically, you can more easily start using products. The second constituent there is the developer. Why do developers love SaaS products? They love SaaS products. They love consuming SaaS products not because of the billing model, but because of the delivery model. The last kind of constituent here is investors and business owners and basically SaaS. SaaS models. SaaS revenue models are just magical for these types of, for investors and for businesses. Why is that? Because they're both recurring and predictable and you can basically predict revenue ahead of time and therefore raise money early on to grow even faster. But when entrepreneurs launch a SaaS business, it's often quite hard to know how they're doing. And that's what we're seeing every day. They're kind of coming to US Stripe as a SaaS business, wondering how good am I doing? How good is my business doing? Are my numbers any good? So this is obviously a pretty difficult question to answer, and I'm sure many of you guys in the room are kind of thinking about that question because it depends highly on the product, on the market, on what the founders want to do with the company, and on the different types of markets you're in. And so nevertheless, there are sort of a few rules of thumbs out there today that I kind of wanted to share and discuss with you today. The first obvious rule of thumb is to identify the standard metrics that are essential to running a SaaS business. Whatever the SaaS business, there actually are a set of metrics. And I'll go pretty quick on these. This is pretty important for companies because getting these right at the very beginning will not only provide you short term growth, but also will have a huge impact on long term given how compounding revenues for SaaS businesses are. So we're going to have a very quick look and I'm going to kind of speed through like the main metrics to then go into what are the levers and the main levers SaaS companies can play with. First one, and the most important one is obviously the mrr. Monthly recurring revenue or annual recurring revenue for some types of companies. This is basically the metric you want to be improving all over this slide. I think when you've understood this slide, it basically structures all the levers that you can play with. With a SaaS business, your monthly recurring revenue changes every month and from month to month it will change based on three things. One is your churn. SaaS businesses have churn. Two is the growth of your existing customers. Three is your new customers. That's kind of how we're going to structure the rest of the discussion today. So let's quickly dive into a few more metrics. Churn. I think we're all familiar with what churn is. It's basically lost customers. And this is a metric that you kind of look at at a monthly or annual basis. So typically a business who loses 5% of his customers every month, you can basically come up with the fact that the average lifetime of a customer for that business is going to be 20 months. Churn defines your average lifetime of your customer quickly on gross margin. Gross margin isn't really something we look at that much in SaaS businesses because it's basically your MRR minus your cost of goods sold. SaaS businesses don't have high cost of goods sold, so they have between 5 and 15%. So we often just kind of talk about MRR. The other interesting thing specificity about SaaS businesses is that you don't really kind of try to optimize for the cogs because you'd much rather work on churn and average revenue and new customers than on reducing your cogs, given how much lever there is on these first three levers. ARPU is another notion. Average revenue per customer. It's basically how much does a customer bring you revenue on average. Now the reality is the real metric you want to look at is how much is a customer going to bring revenue to you over its entire lifetime. And that kind of brings a few, few notions together. We talked about churn. 5% monthly churn gives you a 20 month average lifetime. Well, how much does that customer going to give bring you over 20 months. If he's paying €30 per month, that's going to be €600 lifetime value. That's the main like one of the main metrics you're looking at lifetime value. How much is a customer going to bring you over his life span using your product? This is very linked to the cac. The next one, customer acquisition. Customer acquisition is basically how much do you spend in terms of sales, sales people, sales team, and in terms of marketing to acquire a new customer. LTV and CAC are obviously super important to look at together. If you have low LTVs, customers basically not bringing you basically small amounts of revenue over their lifetime, you can't have a high cac. You have to have a model where CAC is very low and where basically LTV pays for CAC with a certain multiple. That equation basically makes up your profitability. Now talking about profitability for a second, we all know the stories vehicled by the press about the hockey stick kind of growth in the Internet but also in SaaS businesses. The reality is SaaS businesses, more than many others are prone to kind of losing money at the beginning. Like they're working on their product that hopefully is scalable without having much customers. And customers, the few customers they have are paying on a recurring kind of revenue basis. So you're not kind of touching lots of money at the beginning. The good news is given this is predictable and recurring revenue, you can actually raise money pretty early on because you're actually able somewhat to draw out this curve to your investors. So now that we've kind of done a quick run through of the main metrics of SaaS businesses, I'd like to kind of focus on these three things. Basically the three levers of growing your SaaS business, getting more customers, reducing churn and increasing your ARPU per customer. Start with increasing the number of customers, getting more customers. The key here is, is knowing what your sales model is. And there are basically two sales models out there, kind of extreme sales models out there for SaaS businesses. One is what I call the low touch, the low touch sales model. This is products that sell themselves. This is products that kind of sell themselves on channels such as website, email, email. Free trials. These are products that usually have a free trial, then kind of upgrade to pay kind of model. They typically have a relatively low arpu. But also given this is self serve, a very low cac. This is the self serve type businesses. High touch sales on the opposite is where customers need some help in deciding whether or how to adopt your product. This is basically designed around there being an intensive human interaction and human process to convince businesses to adopt your tools, successfully operationalize it and continue using it. This is the Salesforce model. And Salesforce basically invented the playbook here. These are typically high arpu, like high revenue per user, tens or hundreds, sometimes millions of dollars per year, but also generally pretty high CAC customer acquisition costs. And you really need to know where you want to be on kind of that spectrum. There's obviously hybrid models. Companies try to do both and some of them succeed, but it's actually hard because the DNA of the company is usually either on the left side or, well, either low touch, either high touch. Just to say a little bit more about this, if you look at kind of the sales functions are structured low touch, you're going to talk a lot about what we call customer success teams. Customer success teams are basically about building at scale this self serve engine, both in terms of lead generation, in terms of support, in terms of potentially sales. And one of their roles is also going to be to upsell, you know, from a free product to a paid product or to upsell different types of products. High touch sales on the other hand is basically, you know, sales is the heart, the beating heart of the organization with like very specialized sales roles from the SDRs at the qualification stage to kind of, you know, generate leads to account executives who will close those deals and account management who's basically around for retention and also upsell of different products. These are fundamentally different models. Again, there's no rule here, there's no generality, it's hard to make generalities. But generally low touch models have a very, very strong focus in terms of culture and company on the product. Because you need to have a product that's easy enough to use and understand for it to be able to sell itself on its own without any human interaction. High touch sales on the opposite, since they're much more human intensive in the process sales, the sales team is usually a pretty big beating heart of the organization. Again, there's no rule here. That's generally what we see. So there's always this question that we get about how do you get your first customers? You have a great product. There's basically two things here. First is really make sure to have the best product for a given need rather than a good product that fits many needs. That's rule one. Rule two is just building a product won't make your users or customers use it or won't get you any customers in the first place. Anyway, this is kind of the founding story of Stripe. When Patrick and John in 2010 were at Y Combinator and kind of spend their days doing office hours with the whole cohort of Y Combinator and a little water in the Valley, working, sitting down with folks and kind of helping them to integrate Stripe on their new app. And kind of the day after trying to meet with them to kind of have feedback on what they had seen. And that lasted for roughly 6, 912 months. That's what they did for 6, 912 months. They built the product alongside their customers and their users. This is also how Stripe continues to launch new products. We launched a billing product a couple months ago and, and the way we launched that is we identified a certain number of potential users that we thought could be interested. And our product managers work with those users day and night to kind of figure out what exactly we had to build. Some of these users are in the room today and will recognize themselves. And so thanks a lot to them again. So we discussed kind of getting more customers. Next thing is around reducing churn. The rule here is churn will basically make or break your business. And I think the next slide is pretty telling in that regard. If you kind of plot, I know it's small, but if you plot monthly churn 12 months out, you pretty quickly realize, since this is compounding, that if you have a 10% monthly churn, you basically got rid of three quarters of your business over a period of 12 months. If you have 5% monthly churn, you've got rid of half of your business over the first 12 months. And if you have kind of 2% monthly churn, you've gotten rid of what, 25% of your business. Churn numbers can actually kill your business. That's basically the message here. And you don't want to be in that kind of green or even, what is it, orange, yellow category. So what is, you know, after having said that, what is a good churn rate? Again, there's no rules. It depends on, it depends on the product, it depends on the industry, depends on the length of the contract, if there's a contract. It depends on consumer business. It depends on so many things. That being said, there are a few sort of rules of thumbs here that people, investors and others look at. If you look at low touch sales models, think back at what I was saying before. 2% monthly churn is basically what you can, you know, is the best you can hope for. 2% is a very good monthly churn for a low touch type business. 5% is okay, 10% is a problem. 10% basically means you don't have any product market fit and your business is, you know, is not sustainable. For high touch businesses, high touch sales businesses, it's a little different. We generally don't look at monthly churn, we kind of look at annual churn. Given the nature of those businesses that have annual or multi annual contracts with much bigger kind of price Items and tickets. 7% annual churn is good, is actually very good. 10% is kind of reasonable, but you want to be careful if you start going beyond the 10, 15% range. Again, this is annual for these high touches. One last thing on churn is there's kind of this notion of churn is bad. So I should absolutely avoid churn and maybe I should actually make it easier for people not to be able to churn. That's not quite true. So what we see every day working with SaaS businesses and even kind of ourselves is there's a few things here. Again, no silver bullet. These are observations. Requiring credit card details before a free trial usually helps because you get much more qualified kind of people. You can focus more on those. Make it easy to up and downgrade. There's nothing worse than a customer who doesn't have the right plan for him because he's basically going to want to churn and make it easy for him to downgrade. It's always better to have a customer who kind of downgrades than a churning customer. And the last thing is kind of let real quitters quit. Otherwise you're going to enter all sorts of problems. You're going to have chargebacks, you're going to have extra costs because people are going to start to say, no, I didn't want to pay for this service. Bank, please reimburse this charge for me. And then you're going to have as a Merchant, as a SaaS business, so many charges due to that. Now back to the three levers. Now like to talk about the last one, the arpu. And this is also a personal perspective. I believe pricing is not enough. There's not enough investment from companies going into pricing and thinking about pricing. And when you think about it, acquiring new customers is kind of hard. I mean, you got to go into new verticals. You kind of have to do more marketing, you have to do more sales. It's hard. Reducing churn is hard. Reworking your product, everything we just said, changing a price. Yeah, I mean you kind of have to change the website, you have kind of change a few things in the background. That's Something that's kind of relatively immediate, if you want it to be immediate. And you'd be surprised the number of businesses we see, and even us at Stripe, I realize that internally, the number of companies we see out there who at some point when they launched a product, put a price on it, you know, it was relatively rationalized, this is where we're putting that price, etc. And then, you know, three, four years down the line, the price hasn't moved, but the market has moved, competition has moved, the product has moved. So many things have moved, have changed and they haven't really kind of thought about pricing and how that should change. Pricing is so sticky inside companies. And the other thing is the impact of pricing is so huge. Imagine you're able to kind of increase your pricing by say 1%, 10%, that's a 1% or 10% increase of your company value. In a SaaS business, we're looking at pure margin here. So I believe companies kind of don't think enough about pricing and tend to focus much more on the very hard problems. But I think this is like Netflix increased their prices by a dollar, you know, a few, what is it? A few months ago, what did it change for customers? Not much for Netflix, you know, that's a, what is it, 3, 4% increase of value suddenly so talked about pricing. You know, new customers are sensitive about pricing, but there's also your existing customers. That's the beauty of SaaS businesses. And there's so much you can do there to increase your arpu. You can obviously upsell and launch new products and upsell new products, but there's also kind of like your customer base. And some companies have done a good job at kind of working, you know, building products for customers that just grow tremendously. And if the usage of your product inside a given customer grows tremendously, chances are your revenue, your ARPU is going to grow tremendously. So think about the kind of customers you're aiming at your, you're targeting with your product because that alone can be some sort of, you know, amazing rocket ship for your business. I'd like to come back and say just a few words about pricing and kind of pricing models and what are the different pricing strategies. So there's kind of the flat rate pricing that we've all kind of experienced, you know, $30 per month or you know, it could be 10,000 or $50,000 per month. And then you have also the kind of usage based pricing that's typically the AWS model, you know, depending on how much space you take service space, you will pay, you know, more. Those two are great models, but you know, each one of them has their flaw. And typically the usage based model is very good because you know you can adapt your pricing to basically your underlying costs. If you're Amazon, it's kind of, you know, servers space. But for many customers that can be kind of daunting because they don't sometimes don't really know how much they're going to consume. You're launching a new business, you're not quite sure how much you're going to consume in terms of storage space. On the opposite, the flat rate is kind of an advantage because it's kind of predictable for a customer. He knows his cost item, his cost line and his P and L is going to be X. But for the merchant, for if you're Amazon Web Services, you can't really rely on that because your cost might be higher than what you price. So what we're seeing more and more of, and kind of a good practice is this hybrid model where you basically have a flat price but by tiers. So if you consume up to X, you'll pay 100 bucks per month. If you consume between, between X and y, you'll pay 500, et cetera, et cetera. And people like this is something also that surprisingly companies don't do enough and that can actually be helpful in terms of customer acquisition. Another point on pricing strategies is there's the classic that's been around for tens of years per user model pricing. I'll give you a license or whatever it is for, for each user and this is how much it costs per user. The problem here is so many of these clients and customers, the impact, they're buying your product for a reason. It's going to improve their sales, it's going to make something faster internally or externally, more business, et cetera. But that only depends on how well the product is adopted internally inside that customer's company. Right? Like if you're using Salesforce, if nobody actually, if nobody, if you have licenses all over the company but nobody uses Salesforce, you're not getting much out of it. And so this kind of idea of per active user and active, you know, you can define companies spend, you know, hours defining what an active user is, is actually much more reassuring for a customer because it kind of hedges the risk of change management training that are not like easy, simple stuff to do, especially in large companies. One last thing is a quick comment on the fact that one thing we're seeing a lot as well is Companies launching a product and adopting a certain pricing strategy, pricing model and launching another product and they kind of automatically adopt the same type of model and then another product, same thing. If you have a like portfolio of products, you should really think through every single product what's the best way of pricing it for your customer. And sometimes you can also do bundles. But try to think from first principles at that point every time you launch a new product. Because there's nothing worse than a product that's badly priced according to the wrong model just for the wrong reason that your other products are built on that pricing model. So that's, I mean that's kind of something we've been paying a lot of attention to. We have kind of processing, et cetera, that's based on usage. How much do you, you know, do you, how much dollars go do you get paid and you use stripe for. Then we have also some flat rate pricing for other types of products like Atlas that basically enables you to incorporate a company in the US and have, have a bank account in the US as kind of a flat fee once a one off flat fee. And every time we launch a product we really think through the pricing and we have also other types of models that are not on here. So I wanted to kind of take a very quick example of a company who actually published a few numbers a year ago. Buffer Great company. So if you look at the August 2017 column, they have quite a few numbers here to just to highlight a few. Churn rate 5.38%. That's pretty good for a low touch sales model. Business revenue per customer is $17 over the whole lifetime the customer will bring $324. You're not going to build a whole freaking sales team with salespeople spending days to close a customer. If the customer is going to bring you $324 in the matter of its lifetime. And that's why you have a customer acquisition cost of $40 which is basically scaled marketing and things like that. So pretty good multiple between that CAC and that lifetime value ltv. Now interesting thing we talked about like what are the levers to improve your SaaS business? I'm sorry this is a little small. I'll try to voice it over. But basically if you look at a year before same month August, their MRR has gone up by 40%, 43% actually. And if you think about the three levers that we discussed, well one was customers. They did plus 28% on new customers. Churn rate. They increased actually, which is bad. But they increased their churn rate by nearly 6%. But on the opposite, they also increased the revenue per customer by 11%. So overall their MRR has grown by 43%. That's kind of how you think about optimizing and maximizing a SaaS business. So talking about buffer, once you've found your product market fit, you're kind of immediately looking at acquiring new customers. And the beauty about SaaS businesses is that they are possibly the businesses that scale the best internationally. You're building a SaaS product. I mean, there are exceptions. Of course, if you're working in the healthcare industry, there are exceptions all over in the space, but they are generally the businesses that scale best internationally. English is kind of okay as a language for most businesses to work on if your clients are businesses. And so one of the easiest, one of the ways to kind of scale and grow is to go international. And when you go international, getting paid is not just about currencies, it's also about payment methods. And people pay with different payment methods depending on where they are in the world and what types of customers, you know, businesses or consumers, et cetera, they are. And companies don't want to spend too much time thinking about how to solve for this complexity. In the same way they don't think they don't spend enough, they don't spend any time anymore thinking about web hosting and things like that. This is one of the problems we at Stripe have been working on solving. Building economic infrastructure from payments to many different things so that you SaaS businesses don't have to do it yourself. And looking specifically at SaaS businesses, we've built this kind of billing brick which kind of takes into account all the complexities of international growth and recurring revenue and various plans by currency, et cetera, that you guys deal with on a day to day basis. Now why does this matter? It matters because We've just spent 25 minutes talking about SaaS businesses. What are the levers? And I tried to disclaim every time by saying, you know, there's no real rule here, there's kind of rules of thumbs and you guys will need to figure it out. That's true. There's no silver bullet. And the way you kind of recognize fast growing SaaS businesses, the way you become a fast growing SaaS business is you have the flexibility and the agility to test things out and to adapt and to build and to break and to rebuild and to re break. Pricing is one example. I give, I spend a little bit of time on that. And you basically want to have an infrastructure like as a SaaS business that's flexible, agile, that can enable you to do that. Your developers are busy and you should keep them busy on working on your core product. We all know developer talent is scarce. That's one of the rarest resource you guys have. So I'd like to leave you with one question today. In a world where tech talent is so hard to find and to recruit, what are your software engineering software engineers spending their time on? Are they spending their time on building new products, or are they spending their time on technical debt and trying to adapt to all the things you're trying to launch all across the world? Thank you very much.