Amplitude, Inc. (AMPL) Earnings Call Transcript & Summary

Okay. Hello, everyone. And thanks for attending this session UBS's TMT conference. My name is Taylor McGinnis, and I'm the lead analyst covering the SMID-Cap Applications, SaaS and Communications means here on our software team. And in this section, we have Amplitude with Vuong, the CFO. So thanks so much for joining.

Well, thank you for having us. We appreciate it, yes.

Yes, of course. And before I dive in, just for those in the audience, I'm sure you've heard this a million times now. [Operator Instructions]

So maybe we'll start just with an intro on Amplitude, right? There might be some in the audience that aren't as familiar with the story. So if you could provide some color there and even to an evolution of your guys' product more recently.

Sure. Happy to. So Amplitude, we are a digital analytics software company. We provide software to companies like the DoorDash, in Atlassian ad the hubspots of the world that really want to understand their customer journey, and they want to look at it from the self-service capabilities. And if you think about the evolution of analytics, you can -- early days of marketing, it was more of like how many people do we get to our site, how many visitors, what are those -- how many clicks do we get? What was the conversion rate? In the modern day, you kind of actually want to go much deeper than that. You want to kind of start asking like, let's say, you're calm and you want to understand, hey, what causes someone to go from a free subscriber to a paid subscriber? Why did they convert? What activities did they do before that? And so those type of questions are what basically Amplitude really empowers. And in many ways, in the early days, like Facebook and like LinkedIn or Netflix, they kind of built their own because they had to. And then many years later, obviously, companies like DoorDash and others, they say, hey, look, that's how our core competency. And so they kind of turn to in Amplitude to kind of help them answer those questions. And then last year, when we went public as part of our IPO, we actually announced new 2 expansions in our product where we launched out Experiment and what we are now calling Audience, which is part of our CDP offering. So our analytics, we now have experiment and we also power into Audience, which allows people to say, hey, this cohort of customers, let's do something with them. And so you can take that and integrate them and build that into like a braze or an interval or an INTERCOM. And then obviously, Experiment is like AB testing and trying to figure out the results of that.

Perfect. So we're just going to dive into some of the most topical points of conversation, which would be, one, what you guys saw during COVID, right? And now two, what you're seeing now in the tougher macro? So the product analytics market, right, and selling to products team is fairly new, and it's more and more emerging firms like Amplitude and others that paving the way here. So can you talk about your growth rate accelerated all the way to 60% plus, right? Over the last like year or so? And then you've seen growth moderate a little bit more on some of the tougher expansion comps that you've had. So maybe as some history there, could you maybe talk a little bit about what you saw during COVID and some of the tailwinds and what you're lapping now?

Yes, sure. So we obviously benefited a lot from -- a lot of our customers are those digital companies that really grew very robustly in 2021. And as a result of that, just a quick reminder. We charge basically on both the technology you buy or software you buy, which is a platform. And then the main driver behind that is what we call event volume. And event volume, think of it as an activity that a consumer would take on your end product. So if that's like a DoorDash, it would be someone going in and ordering the dinner, those kind of activities. So obviously, as those companies grew their event volume grew with that. And so we obviously had very significant growth and expansion as a result of post-COVID and the acceleration of all those digital. And it's not just obviously in food, but it happened also in payments. It also happened in e-commerce, it happened in even software company, it happened in streaming and media company. We're actually pretty diversified, but obviously, still a lot of digital activities. You have to have digital activities for there to be digital analytics. And so we obviously saw a huge expansion that came from that. That drove our growth, like you said to the 60s in 2021. Obviously, as we go into 2022, the comp on those expansions are actually a lot tougher. And so the expansion on those -- those same customers that actually grew, they're actually still growing. They're just not growing at fast of a pace. And so I think that we're seeing the impact of that slower growth rate from our end customers. Now one of the things that we did actually, announce also in the last quarter though the macro environment is actually causing both some good and bad. I mean on the bad side, we're obviously seeing more churn coming from small businesses. We're seeing some companies getting impacted from their budget cut or their slow growth rate. At the same time, the macro environment is also causing companies to really look at what they're doing. And there's actually a lot of digital kind of native companies that were actually building out their own analytics solution internally and in-house. When basically resources and investments are kind of uncapped, they were like, yes, we can do that with 100 people, and we can do that with 50 people. We can just kind of pile people on to it. As the macro environment has gone tougher, they basically said, wait a minute, is this our core competency? Is this what we should really be building with our resources? And when they look at those ROIs, they're actually churning in looking and say, hey, how do I find a best-of-breed solution. I absolutely still need to understand my customer behavior, so I can build a better product experience. But should I do that with pure man labor or should I do that with my internal systems? And we actually saw that in Q3 because we actually reported our largest land quarter ever because we're also signing up new customers that are in this environment going, hey, I'm actually going to invest in product and my user, but I need to figure out a different way of analyzing and understanding that data. So we're seeing that kind of both sides, the good and the bad from the macro environment.

Yes, that was really helpful. And then maybe focusing in on like the expansion side and we'll talk about the new logo acquisition a little bit later. So earlier in the year, you talked about uncertainty in timing, right? With some of these larger expansions? And obviously, you just talked about the tougher [ comparators ] there. So can you give the group color on what were the bottlenecks with the installed base growth at that time? How conversations have trended since maybe you can touch on the path of existing customers, right? And there's a range that you see. But what that tends to look like, right? And if you guys have made any changes to help facilitate some of those conversations?

Yes, some of our largest expansion in 2021 and it's kind of amazing in [ 2021 ]. We benefited again both large expansion from companies that just naturally grew a lot digitally because of post-COVID, but we also benefit from expansion from customers who said, hey, initially I was working with you only on one product line or one business unit, and I'm not going to expand across my entire business. And we saw that, for instance, in cases like PayPal and Intuit where initially, for instance, we got into PayPal because of Venmo. But now we're in over a dozen different products inside of PayPal. We started initially with Intuit on a very niche new app that they were building for $10.99. And now we're across both all QuickBooks and all of TurboTax. And so those expansions didn't have so much to do with, call it, the increase in business they saw during post-COVID but it's just our adoption and expansion of Amplitude. And so we saw all that kind of happening pretty much in the same time in 2021, which was just kind of like how it naturally put together. As we guide into 2022, the visibility on what we would call vertical expansion, people just increasing the event volume in their core business. That's a lot easier to see because you can look at your data and kind of say, hey, how are they utilizing their existing contract. The one that's always been harder for us, which is, hey, when is a company going to make that decision to standardize on Amplitude as their digital analytics platform for not just one product but all of their products. And so that's something that we were talking about earlier in the year as we're kind of getting into and trying to go. We actually have some really great companies as customer, but we're only in a portion of their journey or a portion of their products and how do we expand that and how do we get that visibility? That was a challenge. What have we done in that. I think obviously, we bought on Thomas in July and some [indiscernible] . I think the biggest thing on those type of customers is obviously just building more relationship at the executive level. We typically come in landing on a smaller dollar size, and we typically have like a product manager as our champion. But to truly get to that level, what we saw in our larger customers is eventually that moves to like a CDO or some kind of data officer, someone that owns data across the company. So building those relationships, getting not just one person, but a pod or a team of folks to really help that customer activate and adopt. So we're doing a lot of work around identifying those opportunities, making sure that we get not just like a customer success person but if it needs a salesperson or needs a solution consultant, how do we get [indiscernible] in there. And really it's about elevating the discussion and trying not to take it just only at the tactical level of that product manager initially. We're trying to bring it into having the overall conversation about, hey, what are you using to understand your customer journey? And don't you care just like you think about the story about Atlassian or Intuit, their biggest value is about attaching and selling other products to existing customers. There's no better way to understand that to connect all those things together. And so that's part of the reason why we've had so much assess in those companies and trying to share that with other companies is a big piece of how we will get better about forecasting those expansions.

Perfect. And then one thing that you mentioned more recently on the last earnings call as it relates to some of these -- as it relates to some of the expansions is the potential for rightsizing, right? In this environment. So I guess, how common are those rightsizing conversations today? And is it possible that the moderation and dollar-based net expansion rate that we could see could be more tied to, hey, maybe companies over extrapolated, right, from some of the digital trends that they saw in '20 to '21 and earlier 2022? And just any color you have on conversations that are trending there?

Yes. So I'd first say that traditionally, companies when they buy in, they typically would buy less than they thought they would need. And that's resulted in then them doing sometime an early renewal with an early expansion. So for us previously, having a discussion of calling it rightsizing or downsizing was rare. I mean never kind of really happened. We obviously started seeing more of that over the last couple of quarters. I would say that it's still a small percentage because people tend to still buy and under buy. But relative to it never happening before, now you're starting to see it. And I think it happens in a couple of flavors. Probably one flavor is that a company for some reason, it's having even a more material impact to their business, and therefore, they're seeing a decrease in their event volume, and they want to have a discussion of that renewal to rightsize their contracts. We do have a few of those. But I think the bigger one or the one that's actually where you have more conversations, and I shouldn't say bigger one, but just more frequency is actually customers where they're actually having internal budget discussions, right? And they're saying, hey, help us understand how we could be more efficient, how we can have better ROI. Now luckily for us, this is something that we actually started working in like 1 year 1.5 years ago because sort of like an AWS, we wanted to increase the level of transparency, we wanted to give people more control over, hey, do you want to bring data in batch or do you want to bring data in real time, which data are you bringing, which days being used the most? And so by giving that visibility to customers, customer can then also say, hey, look, if I have to make this decision, maybe I don't need these type of data to come in. So they can control which data comes in and out and also how it comes in. and giving them more control over their own costs. And obviously, if they're not doing that, that also helped our costs. So it's kind of -- there's some alignment there. But yes, we're having more of the discussions just as kind of the budget happening. Our end goal, though, is to make sure that the customer feels like we're a great partner of this because we know that it's sort of like in 2020, to be honest. I mean, in 2020, right, when COVID started hitting, we had a lot of discussions with customers because the level of uncertainty back in that market, time frame was actually fairly high, too. And I think we made the right decisions with those customers to either take a partial churn or work with them on whatever the contract they needs may be. And it eventually led to a really amazing 2021 with a lot of those customers. And so I think we look at the long-term perspective of these customers as long as they are invested in digital analytics and in Amplitude. What we may have to do with them in the short-term to make sure that they're happy and they're satisfied seems to be the right moves.

Got it. And then I think tricky part is understanding like sensitivity of some of these models that are more consumption/subscription models, right, that are tied to digital activity just because we're living kind of an abnormal environment right now, right? So you saw a huge uptick of digital activity in the pandemic. Now we're getting in a tougher environment, and I think there's less visibility to what happens to a lot of that digital activity as things get tougher. So is there any examples, I guess, or anecdotes that you can provide that would say, hey, here were some of the tailwinds to digital activity, right? And what some of your right have seen. And then to the extent you're having -- when you're having these rightsizing questions with our conversations with customers, what are being brought up as some of the headwinds, right, that we're seeing to digital activity that might create more of that sensitivity?

Yes. I mean, I think that in general, when we look at it even like our -- what we call event volume usage, what we see is that it's still growing year-over-year. And most of our customers are still growing is a question of their growth rate. And so the small number of customers that actually are seeing decline in activity, they tend to be in very specific either vertical or kind of stage of company, right? We're not really seeing that, call it, decline at the larger enterprise or even more established, whether it be consumer Internet or B2B, like in the growth rate may slow down, but they're -- the overall on volume of digital thing is still growing. Obviously, for us, like given obviously when we saw a pretty big increase in cryptocurrency in 2021. That one is a clear one where there's definitely a decrease in activity in 2022. Same thing would be maybe, say, early stage start-up. You saw a lot more of those companies coming out and started growing. And then this is a little bit less. And so they tend to be more focused on certain verticals where we're seeing that. But like for the most part, like if you think about a real example, I would say most of them that are no material customers, they're still growing to smaller. And then anything that we're seeing less broke, we're also seeing counterbalance to that. So for instance, in 2020, even in 2021 -- well, before 2020, like travel stuff like that because it's so big online, but actually a big, big portion of our business. Obviously, it became a lot smaller in 2020 because if you think about it -- I mean, that one was very abrupt. I mean like the shutdown on those activity was really pretty stark when the whole world kind of basically stayed at home and now that is actually really kind of expanded. And so I think that we continue to see that. And I think that's one of the benefits of the fact of being actually pretty diversified across different use cases on digital. And so the big question for us is always like, do you believe that overall digital activity, whether they it be travel or whether they it be streaming or whether they it be gaming or whether they productivity B2B. You think that that's going to increase in the long run. And obviously, our bet is absolutely, we think it will. And so that's why we feel it's okay. But there will definitely be ebbs and flows by vertical.

Yes. And then maybe as a follow-up question to the sensitivity, right, of the business model. Anything you can provide on how like the pricing model ties, right, to digital activity? And I think you talked or touched on this a little bit earlier but how easy is it to control event volumes if a customer wants to save on spend? I don't know if you mentioned that might be something you are working with customers a little bit more. But can you talk about that?

Yes. So first, let's talk based on pricing. We obviously -- one of our primary goal is to try to make sure that we're providing a solution that can scale with the customers. And so as they grow, the more you're giving us, obviously, the cost part, what we call million events will go down. Now there does reach a point when you're as super size scale where the rate of decline will be a lot lower because you're actually already kind of at the point of efficiency. And so we do provide that kind of scale for our customers so that as they add more product, they can get better efficiency at better cost. I think we -- it's not only for them but to ourselves to constantly actually increase both the transparency of the data and also the mechanism for them to control that data. And I think over the next couple of years, we'll actually see more and more people have those controls. For instance, I mentioned it earlier, but there's a big difference in cost between real-time and batch. Now there's always been a big difference in for instance, if you bring data whether you sample data, you bring all data. What we've learned is that for most companies that really want to be super successful, they're not going to sample data. They do want to bring in all the different data because they need to understand all their customer behavior and break those in cohort. So the sampling versus non-sampling is a much harder choice. But the decision to do batch versus real time really depends on the use cases of that data. And more and more customers will be willing to say, hey, I can't mention the customer's name, but there's a pretty large customer of our where when they were looking through their stuff, they're like, we think we could actually identify almost 3/4 of activities that could actually go to batch. And that will allow them to obviously take advantage of a much lower cost structure. And so not only that, but also, like you said, they can look at the events that are coming in and either consolidate events. Sometimes there's a cleanup that can happen because you can -- you can buy in a bunch of events into one. Part of all of that is clearly good [ health hygiene ]. I think like I often -- we internally talk a lot about like it's just like the same practice as you try to scale up an AWS, right, between about do you do -- what reserve instances do you invest on versus on-demand versus real, which data transfer you do, what type of thing -- all those optimization things because we're pretty excited that we have so many large customers, so many customers over 1 million plus. But that also means that you want to make sure that those customers have a good sense of control and also really good visibility to how they're getting ROI. And so we don't have any issues out because we think, again, in the long-term, that's a win-win for us. And so we're going to keep adding visibility into how you control your data, which data is being used, which one you can turn off and allowing people to actually again, manage those environments, especially in this type of macro environment, I think that just basically makes it for a better relationship.

Yes. And then on that point of batch versus real time. Are you able to quantify how much is real time versus batch, of your business today and how that might trend near term versus long term, you guys on that front?

Well, that's -- that's easier because like almost all of our stuff is real time so that's how we started. And so doing batch will be something that is relatively new for us. It's something that like again, that was being requested customers as a way to help kind of make sure they could have more cost control. And so I think that we're going to start seeing that rolled out this next coming year. And I think it all really depends on the type of customers. I think ultimately for us, we want to be indifferent about which way they want to go. It's which everyone works best for them because, obviously, if you're going through batch, it's also will cost you guys, but it will also cost us less too.

Yes. And then just on the pricing front, right? You guys price in, I believe, USD internationally, right? Obviously, those customers are seeing big changes in FX rates. So that, combined with everything that you just spoke about now and maybe some of these batch use cases coming to like, is there anything you guys are evaluating on the like pricing side, right? Or any changes you're thinking of making there?

I mean we're obviously monitoring the FX impact, I mean, like as much as it's gone down the last couple of weeks have also had been a pretty wild swing going back the other direction. And so the whiplashes that are happening there, we don't want to tend to -- try to overreact to one way or another. I think that in the long-term, like most large companies, as they kind of scale on stuff like that, we'll get to a point where we'll actually also be dealing with it in local currency. And so then you'll end up taking on that FX risk like a lot of large companies like a Microsoft. Right now, we just didn't do that because of how we got established. It kind of helps us in the near-term from that standpoint, but it also hurts us from doing business with customer right looking at that price list and are going, wow. It's a lot more expensive than it was a few weeks ago. And so I think we're constantly evaluating that, but trying to guess where the FX market is going to go. It's like a lot of other market predictor. So I think for us, it's really about trying to make sure that, hey, ultimately, we will want to enable business in those current country to do it in local just like we do our expenses and then the FX will be the FX.

Got it. And then last question on some of the expansion stuff and take us a little bit -- a little bit away from that. But -- so on the expansion. So when we think about dollar-based and expansion rate, right? Putting all together what we just spoke about, how does that influence how you think about that metric near-term, right, and longer term. And I know it's probably not a great comparison, but is there any -- is there anything we can draw up what you guys saw at the start of the pandemic, right? And any potential troughs, right, for that metric?

Yes. I mean I think we mentioned in the last earnings call. We do see net retention rate declining because obviously, we track as a 12-month thing. And we also mentioned that like we look back at 2020, our net retention rate was like 116%. We obviously elevated and grew that all the way up to like 126% at a peak during -- post-COVID. And so right now, I think the last quarter, we put out was 123%. And so I think that we expect there to be some decline in that just as we kind of see obviously lower expansion and some more of the churn that's happening from the macro. We don't forecast that out, but I do think that number will have some pressure on in the next few quarters.

Perfect. And then I'll go back to some revenue questions later. But I don't think Amplitude was alone, right? And seeing some of the pressure, right, to I think there are other [indiscernible] office companies that saw something similar. But what was actually interesting as there was a lot of strength across the board in new logo activity, right? And I know you guys were one of those companies that flagged a record quarter of a new logo activity. So can you maybe just talk about what you guys are seeing on that front, right? Maybe you can talk about the ROI, right, and your conversations land a customer and how durable that could be as we go into next year?

Yes. I think I mean, it's interesting, one of the reasons why we actually went public was actually we also wanted to be able to increase the level of awareness of both digital analytics and Amplitude. And so I think that one of the big challenges that we have out there and even customers that signed we talked about is that they didn't even know that Amplitude existed out there. They didn't know the solution now they existed for what they were actually doing internally or building internally. I think that the macro environment, like I said, what we've seen from that, there's actually a few things going on this year that's kind of helping us. Number 1 is, at least from a new land, there are -- but the macro environment is actually helping us from a new land by forcing companies to really look at how they're deploying resources and whether or not that's the best use of resource. I think I mentioned that earlier, so that's true. I think the other thing is Google Analytics. They obviously announced that they're moving to GA4. And then for that, they're really adopting in event, new customer event-based schema, which is what we've been doing for the last 10 years. And so -- but when you migrate to the new Google version, you basically lose all of your data, lose all your legacy. And so some companies are like, wait a minute, if we're going to do this migration, like let's look at this, what's the best solution out there. And there's been a few notable wins were -- they were actually -- it was started because they were looking at the new version of Google Analytics and trying to figure out when they're going to do the migration or what the impact of that was. And so we're seeing an increase in pickup of land, both from the macro environment causing people to really focus on their core competency and their resource and also by Google Analytics and they're kind of push up like this event schema which goes right up our alley and people the looking at Amplitude and going, wow, I had no idea that you guys were out there and could do those things. And so I think that's really helping us from the land side with -- especially with companies that already have a pretty sizable digital footprint and have been doing something else to solve their questions around customer behavior.

Yes. I think that product cycle, one is interesting. So in terms of the number of leads that are coming into you guys, right, that our existing Google Analytics customers, right, that maybe you're starting to have to make that shift. Can you provide color on how big of like a tailwind that is becoming for you guys? I don't know if internally, you guys have taken a shot in terms like what that opportunity, right, for Google Analytics potentially migrating to Amplitude could look like. But any color there?

Well, I think that like for us, getting customers that were on Google Analytics has always been there. And in fact, if we go back pre this move, we saw -- there's a question and an opportunity to ask like what's the current incumbent solution or solution someone's using. And we see Google Analytics, call it in the high-teens before. Now I think in the last quarter, we put out, we had to do was about like 25%. So there's been an increase I don't know if that's more of like our own salespeople awareness or that's -- because I guess I find it hard to believe that Google wasn't in a lot of these places already, personally, just because I used to run SEO and SEM, myself, so I'm kind of like, I would see why you want to run Google and why you still need it. And so I think that for us, it's more like the awareness of the advanced-based schema. I think that -- I think it's more of like a progression to be honest, which is companies will either stay, realize, hey, what we're doing building in-house or what we were using as analytics, what we would call marketing analytics to really get to the next generation of truly understanding your behavior. There's only one way to do that, and that is through the digital analytics platform. And then we start looking into that, you realize that Amplitude actually just the best solution out there and so I think is that level of awareness and that level of understanding that like, hey, understanding -- and we talk about internally. So the way we described it as we think about people thinking from novus to intermediate to advanced. And Novus is kind of like people going, hey, how many MAUs or DAUs I have? How many clicks did I have? And then intermediate start people were going, oh, I want to understand my conversion funnel. I want to understand like what group went from this stage to this stage and how do I break that into cohorts. And then Level 3, which is where we actually are the strongest and strengthen its kind of going, okay, great. Well, but why? And what do I do about that? And what action do I take from those insights? And so it's elevating the discussion. And we do this a lot of times. We go into a customer and our pitches typically what question do you wish you could answer now than you can't. And let us show you how you can answer that question inside of Amplitude. And so I think it's elevating it not necessarily just like we're replacement for Google Analytics, it's more of like, hey, as a company becomes more mature in their priority development and their growth cycle and all that, I need to understand their customer behavior better. The insights that we provide, it's just there's nothing else out there that allows them to do that. And most of the most sophisticated companies that we think about like the Facebook and those -- they built their own. In the modern day like most of them aren't going to do that anymore.

Yes. And you talked a lot about raising awareness, right, for what Amplitude does. Can you give us an update on -- and now that you've been public for a year, maybe you can give us an update on what your customer base looks today, right, in terms of vertical mix? Is it really mostly a lot of those digital first-type companies? And are you starting to see some of those more traditional enterprises adopt your guys' solution at scale?

Yes. And I think this is the word or comment that we made where -- again, [indiscernible] like maybe the mistake we made last year, we thought that, hey, we got a lot of digital great companies and we're ready to move into what we call the early majority, start going after the next group of customers that may -- they're just getting into digital trying to transform their business into digital. And then we look at it now, we're now at 1,900 customers. But if you look at like the revenue, the majority of the revenue still come from companies where the majority of their business is digital. Now it doesn't mean that it's concentrated it is basically across all kinds of different, it could be travel, like I say, it could be entertainment. It could be a B2B enterprise business. It can be consumer Internet, it can be financials and payments. So it goes across all the major like retail financing health care, all that. But they're still -- digital is their core primary product of business, and that still is the case. And I think what we saw this year was that actually, there's still a lot of those type of companies that haven't built -- that haven't really gone on to Amplitude or haven't like got the right solution for digital analytics. And so going after those is actually more fertile than going after some of the earlier stage. We still land them. We still land some of the early ones, but their logos may be great, and they look great but their digital footprint is still so small that the impact on ARR revenue smaller until they really expand. There's only like a few stories like we talk about like Ford and stuff like that where their growth is really big, and they've kind of really gone much wider into that digital whereas most of the real -- the digital company is like, I think 2 quarters that we announced like, for instance, Groupon and last quarter, we announced Zillow. And then there's others that we can't mention, but if you think about Zillow for instance, those aren't companies that are new to digital. I mean they started in digital, they're just new to using like an Amplitude for digital analytics.

Got it. And then I got a question from the audience on margins. So maybe I'll shift to that, right? And that's just wanting to understand how you're thinking about top line growth rate versus margin trade-off on your trajectory towards breakeven, right? And I think you have a lot of unprofitable software companies that are accelerating that path, right, given some of the uncertainty in the environment. So curious like how do you guys think about that near-term, maybe even longer term? And are there any initiatives in place to potentially pull some of those levers to accelerate that path?

Yes. I mean let me kind of talk a little bit about our journey there a little bit. I think when I first joined the company, our growth margins we're in the low 60s. Last quarter, we reported our gross margin at 74%. Last year, even when we had capital growth in the 60% year-over-year, it was only when we went public, we were trying to argue with the SEC about like whether or not the direct listing costs should be part of your free cash flow. Ultimately, we said fine that they really want to be part of SEF. But if you back out the banking fees and the listing fees and all that from our free cash flow, we actually had like single-digit negative free cash flow while growing at a very great clip last year. And so our efficiency has actually always been there. And I think even this year, we continue to really be mindful of that. I think last quarter, we reported negative 7% operating margin. Our free cash flow will kind of vary quarter-over-quarter because we -- because of our billings and kind of when the renewal base is. Now for instance, most renewal base, our largest renewal base is actually in Q2, right? And so we tend to have a much stronger free cash flow like reported positive free cash flow in Q2 of this year. And so I think as we look out to next year and beyond, what we said earlier in this year was that we are aiming cash flow breakeven to positive in the medium-term. And we define that medium-term as 3 to 5 years. We obviously are passing a 1-year mark of that. We are still well on track to make sure that we do that. And so we're going to stay focused on making sure that we stay efficient, get to that positive free cash flow. We're also very mindful of trying to make sure that we're following a rule of 40. Obviously, you're not going to be able to hit it all every year and but like how do you make sure that, that's something that you're being very mindful of. And then that -- those principles will then controls how much we decide to invest given the macro environment and all that. And so we saw -- like, for instance, this year, we were at 19% of revenue for peak part of the development last year. We're now at 22%. We're actually very happy to do that because we felt like we had strong condition in our product road map and the new products we're launching, we can kind of level that off now. We don't need to keep driving that higher. G&A will be something that natural would just get more efficient because you ramp that up as you're going public. And so by far, the biggest lever, obviously, is go to market. And I think -- I wish I had a perfect crystal ball to kind of know exactly what's going to happen in the future. But we want to try to invest as much as we can because we believe the market and our product positioning while trying to keep our first principle in mind. So that's how we try to offline. We go -- what's the maximum we can actually invest because we know that we want to make those bets for the future in our market and our product while being mindful of not getting ahead of our skis. Now I think when I look at some of our peers, I kind of chuckled with some of our Board members because last year, they were beating me up they were saying that we were too efficient and we were too conservative and all this stuff. And now this year, I get a lot of calls from my buddies who are obviously, they're going -- so and so keep calling me to look at you guys. You guys aren't huge scale. You guys aren't like the multibillion, but you guys are actually so efficient. Like how do you guys do that? And so I think that's just in our core DNA.

Yes. And then on that, so when you're thinking about some of the cash flow targets that you laid out and how operating income margins might trend, right, in conjunction with that. Is it fair to say that if you take out some of the costs of going public, right, things that were more onetime in cash flow, that the difference between those should stay steady, right, over time, such that the view of the company is that you'll continue to show leverage in those metrics? Or is there anything that we should keep in mind over the next year, the macro weekends where you could see volatility, right, in that path?

I think that we have enough control over our lever that you should expect to see that where we want to kind of show continuous improvement. And obviously, if the macro really picks up, and I think see that growth rate is going to really pick up, we may actually increase our investment in spending. So the operating margin may be a little bit worse, but that's going to be made up by growth. If the growth really kind of comes back and isn't there, we'll dial back the investment on the go-to-market side and other areas to commensurate with that. So I think that we look at them in conjunction. We don't think of it like necessarily isolated.

Got it. Perfect. And then I know you don't like the billings question. But it is one that gets brought up...

Yes, Taylor. Yes. Love billing, love billing. I love billing, I do like it, that's how you can get caught. I love billing, I don't like the billing questions now keep going.

Yes, yes. yes. So I saved for the last one. But there obviously has been some different [indiscernible] into cash flow, which we were just talking about. But there's been some volatility, right, on the billings line. Obviously, that function of renewal. There's also been some different trends on the CRPO line, right? And just how all of that ties into what you're seeing in terms of like revenue growth. There's been a bit of a divergence there. So maybe you can just talk about the impact to each of those lines, why that might be differing from revenue growth or how we think about even those metrics in the medium-term.

Yes. So I mean I assume back on, obviously, everyone is trying to figure out what's the best leading indicator to look at our business. I would say the best leading indicator is to look at our revenue growth because there are different noises both in our billing and CRPO and I'll talk to a little bit that the answer to your question. On the billing side, what -- there's a few factors that makes it interesting for Amplitude. Number one, it goes back to when customers actually do expand and when they do renew. We tend to -- like I said, they tend to -- a lot of time under buy and then, therefore, they expand early. And when they expand early, that actually starts a new contract for them. And so the renewal base is moving. So some of that, you can't look at like prior year and go, hey, that renewal base is there, and they're going to just add a few more seat licenses and the cold term to that same renewal base. A lot of times, they've already consumed all their event volume. And let's say they were post in Q3 or Q4, they've consumed it, now they need to start a new contract in Q2. When that contract renews and all of a sudden, that billing happens, but that billing happens not only -- it happens for the whole new contract. So you're shifting the building from one quarter to another. So that phenomenon of early expansion and billing it causes us a little bit more volatility and noise quarters and costs just go, wait a minute, was that Q4 or is it now Q2? And so you're constantly trying to play that? The second one is that we tend to -- and this is just historical. We obviously could change it, but we just haven't because it's not something that like we want to introduce a lot of nightmares and everything. But wherever we book or wherever we renew, we tend to bill the following month, which is also not typical. It's atypical because people typically bill in that same month. We bill the month after. And so those 2 things are what causes the billing to kind of move where someone is like, man, I didn't expect your Q2 to be so strong in positive free cash flow. Man,why is your Q3 year-over-year, it's down so much? It's really about the renewal base and that kind of stuff. The RPO and CRPO, the biggest driver there is that we made a conscious effort as part of when we were going public, one of the things that we try to compare against our peers is what percentage of our deal is our multiyear, and we were like very little, like single digits back then. And it was like, well, that doesn't -- why are we so often? So we made a conscious effort to increase to multiyear. Now in hindsight, that seemed to be a really good decision given the macro environment because it actually gives us some more higher level of certainty. It also gives us more understanding of everything and so. But that driving a more multiyear will have an impact on your CRPO because, again, just a quick explanation of it is like your CRPO is [ West ] in the next 12 month. So if you were doing a 1-year contract, if you are 6 months into it, the only thing left in your CRPO is 6 months' worth. But if that customer is a 2-year contract, 6 months into it, they still have 12 months. So you're not going to get to that 6 months until 18 months into the contract, right? And so that is actually what's causing that CRPO dynamics, too. So hopefully, that answers the questions on both billing and CRPO.

Yes, that was perfect. Awesome. Well, I think that's all the time that we had. So everyone in the audience, thanks for attending. And yes, thank you. This was great.

Thank you, Taylor for having us. Appreciate it.

Yes, of course.