MongoDB, Inc. (MDB) Earnings Call Transcript & Summary

Okay. Good morning, everybody. My name is Tyler Radke. I cover the Data and Analytics and vertical software space here at Citi. And for our next presentation, we are happy to have MongoDB. We have Michael Gordon, the CFO and COO; and Serge Tanjga, who is the Vice President of Finance and helps out in Investor Relations. So gentlemen, thanks for joining us. Normally, we're doing this conference in New York, which is right at your doorstep, but I appreciate you joining in the virtual format this year.

Yes. Happy to. Thanks for having us.

Thanks for having us.

So Michael, I thought we could just start off with the topical topic of your recent results last week. And I thought there are a few things that stood out to me. One, was just kind of the overall commentary on the demand environment, specifically what you're seeing within your cloud or Atlas business in terms of improving expansion trends. Maybe just give us a sense for kind of the key puts and takes on the quarter and what you saw from an improving business trend? And how that's kind of played out since you closed the quarter?

Sure. Yes. No, thank you. Happy to. I'll just take a quick step back and talk a little about the quarter to put it in context for folks and then jump into a couple of those different areas. So overall, we're really pleased with the quarter. Another strong quarter for us. Subscription revenue growth, 41%. Atlas, which is our database-as-a-service offering, up 66-plus percent. And Atlas, which was introduced about 3.5, 4 years ago, now represents 44% of revenue. So it's been great to see that. And then in addition, we're pursuing one of the largest markets in software, and you saw that flow through the numbers that we had a record quarter in terms of the number of customer additions ending at over 20,200 customers. And so generally, a really strong quarter. Obviously, COVID is an impact. It's certainly a headwind given the challenging environment, but it was less of an impact than we expected. So really proud of how the team executed. The team executed really well and really strong customer engagement, sort of the value proposition is resonating, very large market. But certainly, it's a very challenging backdrop, and that certainly colors our view as we talk about second half outlook. As it relates to Atlas, what you asked about specifically, what I would say is we saw a rebound in improvement in the expansion trends what we talked about in Q1. For those who weren't focusing or don't remember, it feels like eon ago. What we saw in Q1 was roughly concurrent with the global synchronized shutdown of all business operations was we saw that our cohorts of Atlas customers were expanding, but expanding at less rapid rates compared to the historical expansion rate. And so when you dug into it, we were continuing to add new customers and obviously, in Q2 at a record pace. We were not seeing any increase in customer churn, which sort of underscores the mission-critical nature of the database and what we provide. But commensurate with an overall slowdown in business activity, we did see a slowdown in consumption activity, and those cohorts didn't grow as quickly. It also makes sense when you think about the revenue model for Atlas, that those revenue models, consumption-based revenue model. And so it would make sense that things would -- people will be consuming less and there's less business activity. We did see an improvement roughly in the middle of the quarter as economies and geographies started to open up more around the world. We did see an improvement and an increase in those expansion rates, not quite back to historic levels, but improved from the modest reduction that we've seen in Q1. And so that was great to see. And really sort of excited by the improvement. And I think, importantly, it underscores the sort of broad-based nature of the product that we offer cuts across industries, cuts across geographies, cuts across verticals, and the fact that we saw the sort of slower expansion, roughly concurrent with the shutdown and an improvement, corresponding with the reopenings sort of underscore this is really a macro transitionary factor as opposed to something more structural. And again, it makes sense with the rev rec model and makes sense given that we're a broad general purpose database.

Got it. One other thing that stood out in the quarter from my perspective was the record net adds in, I think, both total customers and then Atlas. And that also followed on a really strong Q1 in terms of net adds despite the weaker expansions. One question that we get a lot is just from investors trying to understand kind of the different cohorts or customer types within Atlas. I mean, obviously, you have almost 19,000 unique customers accounting for almost $250 million of annualized revenue. I guess, what is that base of customers kind of look like? And how would you just kind of explain some of the record net adds that you saw in the quarter? And what seems like a pretty difficult environment in software for adding new logos?

Yes. So a few different things. I think that adding a new logos really is reflecting the breadth, the size of the market opportunity that we're going after. It's just a very, very large market. Combined with the fact that we're getting better at some of the self-serve dynamics that we have been talking about for the last several quarters about some of the investments that we've been making. And so I think, overall, the customer count is a great sign. I think we often think about -- I know we report Atlas. And to your point, Atlas is almost $250 million annualized business growing at 65-plus percent growth rates. And so I understand why people focus on that and why that draws attention. But we really think about the business on a channel basis, right? And so we think about the self-service channel versus the direct sales channels. And then the direct sales channel, you can further bifurcate into more of an inside sales, kind of big market team and an enterprise field team. And we've seen customer success really across the board, particularly for Atlas across all those areas. So self-serve, we've been talking about for some time. And that's really -- we've been making investments in customer acquisition, funnel optimization, et cetera, et cetera. And we're starting to see the payoff from those. Again, we've got just -- it's a fact it's just the incredibly large thing about the sort of 125 million plus downloads that we have just from our website alone. The popularity of MongoDB the fact that MongoDB is consistently the #1 database that developers love and most want to work with. And that creates a huge opportunity. Now that we have this database-as-a-service offering, we need to make sure we're increasing awareness of it and successfully driving registrations and converting those people to becoming paying customers. And so we've made investments over the last several quarters in that, and we're getting better at that. We're certainly not done. We've got more work to do. We just announced recently that we just added a Chief Marketing Officer. And so looking forward to Rishi joining the team and incrementally helping us build out that skill set as we've been adding some of that DNA and capability internally in the team. But I think you're starting to see that, the manifestations of some of that show up. Secondly, if you think about the direct sales side, we've made changes or improvements in our go-to market, we've done some sort of refinements and some optimizations, whereby historically, particularly for Atlas, we were really focused on getting and maximizing that initial customer commitment. And that both creates sort of a tension and a friction in the conversation. And particularly for an Atlas workload, which is more likely a new workload. You don't really even know how much you're going to consume, right? So if you're a potential customer, and I've got a sales rep who's trying to get a large commitment out of me, I might be hesitant, I might be concerned. And so what this does is we sort of reduce that tension. We remove that friction. And we've said, look, we have the cohort data. We know that when we acquire Atlas customers they, on balance of the cohort, grow at like very, very healthy rates. And so the most important thing to do is to make it easy for someone to sort of adopt and start using the platform, and then they'll grow and expand. And we've kind of seen that consistently. And so we made some of these tweaks and refinements to the sales team and what it's having is it's having the intended impact, which is increasing velocity. You can see that in the number of customers, not just total customers and not just a subset of certain customers, but you can also see that with the increase in the number of direct customers. And so I think it's been satisfying to -- when you turn or adjust the dial and you hope or expect it to have a certain outcome, it's been great to sort of see that reinforced operationally. The last thing that I'd say away from the day-to-day execution is we take a macro step back when you think about the backdrop of COVID-19 and everything else. All of this mandatory work-from-home, et cetera, et cetera, is only reinforcing all these trends that we were beneficiaries of in the beginning or in the first place, right? And so separate from the fact that it's a huge market and everything else, we're really seeing companies saying, I need to innovate more quickly. I need to transform my business digitally. I need to move to the cloud. I need something that's multi-cloud. And so all these things are sort of providing incremental tailwinds and accelerating some of those trends that sort of have been there for the last several years, but this is a more crystallizing event for a lot of companies. And so you see that in the nature of the dialogue and the customer engagement around it.

I see. So it sounds like there's -- it's a combination of just good execution, building out the self-service, making the tweaks to the direct sales side, increasing the velocity, but also maybe a beneficiary of some of the COVID-specific phenomenon around work-from-home that's driving this?

So I think that's right. I don't think it's not quite as linked as you'd see for a Zoom or [indiscernible] if that's more specific. But I think the overall organizational imperative, right, to adjust and adapt your legacy technology to something more modern, that's more dynamic, more agile, more scalable, more distributed, all those kinds of things all play to our favor. So it's not like I'd specifically point to this set of customers, and we added this many incremental customers because of that tailwind. But I think it's more something that sets us up for the longer run to continue to be well positioned.

I see. Okay. And then talking on the kind of actually -- staying on the execution front and specifically around sales efficiency. I think you have called out good sales efficiency and good sales execution in the last few quarters. And it looks like you're hiring at least in sales and marketing from the numbers that we can see has been pretty robust. I think, in the last quarter, your sales and marketing headcount grew roughly 60% year-over-year. And while I think everyone would agree growth has been strong. I mean revenue growth is a bit below that. So I guess like how should we think about the context of the comments around improved sales efficiency? And I guess if you're growing in the go-to-market headcount, if you will, around 60%, like why wouldn't -- you are seeing that efficiency, like why wouldn't we see kind of the revenue growth kind of follow that at some point?

Yes. Sure. So a couple of thoughts. And obviously, Serge, feel free to chip in here. But I think the key thing, and I think we have some disclosure in the Q around this to try and help people understand this, is the overall headcount growth I understand the natural tendency to presume that, that is proportionate when you think about quota-carrying. I think we've tried to add incremental disclosure to make that clear that we have historically -- if you look at it like on a year-over-year basis, disproportionately added non quota-carrying folks, particularly as we build out the sales and marketing, really marketing self-serve component or aspect of the business to drive that channel. Because we haven't had that and because that's the DNA that we're adding. We've been adding people sort of disproportionately there in other non kind of quota-carrying roles. And so I think the quota-carrying capacity is growing less than that. And in general, there -- the only other thing that I would sort of add if you kind of think through the models is there is a ramp time for heads to be productive, right? And so we've generally seeing increases in productivity, which means that you're getting more productivity for a ramped person, but that doesn't mean that you're getting more productivity for a total person, right? Because you've got lots of -- you saw a high percentage of new people and a high percentage of people who are still ramping, which over time will add and accrete to the ramp productive capacity, but in a snapshot probably gives, especially in a high-growth business that's adding a lot of new heads. It disproportionately a view that's sort of too high relative to what the ramp number looks like. Serge, I don't know if there's anything that I missed or you might want to add?

No, that's right. That's right. The only thing that I would add is the comments that we made around sales efficiency, particularly around the last couple of quarters, were really just to highlight what we believe a strong execution in an unprecedented environment. So in Q1, we were happy with our sales efficiency, and we talked about how Q1 was this tale of 2 halves, right, where the first half of the quarter was normal, and we were generating pipeline and shepherding deals through the process as we would normally. Then we went into a lockdown in the virtual world, and we were very happy with how the team responded. And then at that time, we were looking at Q2, where it's going to be the first fully virtual quarter and what impact would that have on sales efficiency. And we were very happy with the execution and the sales efficiency was better, obviously, than we forecast. That's why the numbers ended up -- part of the reason why the numbers exceeded our guidance. But that being said, we fully believe that efficiency would have been higher and new business would have been even better even in the world without COVID.

I see. So just going back to the ramp commentary. So it sounds like you're fully ramped sales capacity, that is where you're seeing some kind of incremental efficiencies that maybe you're adding -- maybe that's offset by kind of new hires that aren't -- that obviously aren't fully ramped out of the gate. I guess as you think about -- obviously, there's a lot of challenges in a virtual world. And when you are onboarding new sales folks, have you noticed any changes to just like how fast the time to get them productive is? Is there -- is that elongated, just given the challenges of work from home and you're not like in the office learning? And just be curious how that's looked since you've kind of been in this virtual world?

Yes. So given ramp periods, we don't have a ton of data, but it is a thing that we're paying very close attention to for all the reasons that you're describing, right? You're talking about this both in terms of existing reps as well as new reps, right? So it's existing reps, Q2 was our first full quarter of being remote, right? And just as Serge and I both commented earlier, we've been incredibly pleased with how well the team has performed just sort of despite that challenging backdrop. But I don't think we want to get carried away and just assume that. Therefore, we did it, and that's all fine. I mean, I think what people miss sometimes is the environment is incredibly dynamic, not just the situation with COVID, but in general. And I think I'm incredibly proud of how the team has continued to adapt through that environment. But that doesn't mean that there's, all of a sudden, this sudden forget a playbook that exists for like how do you execute well from a sales perspective in a pandemic, right? Like that's not some sort of like easy thing that you can just quantify and stamp out and have everyone turn to Page 32 of the manual when they encounter whatever challenge XYZ is. So we'll continue to stay vigilant and do that. I think the -- one of the areas that is uncertain and that it's just -- it's not possible to have enough data is, is you don't know how a fully remote environment to your questions is going to affect someone 12 months into the job when they had a entirely remote onboarding experience and everything else. We've tried to adjust and make investments and address that and sort of compensate or make offsetting adjustments, but we won't have the data on that for some time, and it's an important thing that we pay attention to it operationally.

Yes. Okay. Totally. Well, I think in Q4, which obviously, you -- I think most of that pipeline that the strong Q4 that you had was generated pre-COVID. But it was a notable quarter in terms of the $1 million-plus customer relationships that you had, and I think that grew over 60%, if I remember correctly. I guess a couple of things there. You did call out in the last quarter, just kind of a tough compare for Q4 of this year on the EA side. Just kind of curious how your large deal pipeline has kind of evolved through the year? I know it's probably like too early to have a super strong view on Q4, but just kind of curious how that pipeline generation activity is. And then in the $1 million-plus customers, maybe just talk about what Mongo looks like relative to other database vendors? When you get into that type of territory, are you typically consolidating different database technologies? Or is that still maybe at a -- is that still kind of out in the future, maybe a $5 million or $10 million relationship?

So why don't I start with the $1 million part and then circle back to Q4 because they're related, but they're also separate. So the way we usually see a customer -- and this won't be a surprise to you, Tyler. But the way we usually see a customer get above the $1 million mark is after several workloads, right? So we have a lending and expand strategy with the -- first, you sell the first workload. And then you look for other incremental workloads and use cases that you can attach yourself to in the customer and you grow that way, both through the growth of the existing workload, but more importantly, through adding additional workloads. And we've talked about how even some of our largest customers, we still have relatively small share. And so crossing the $1 million mark, is sort of a step in that journey in the context of we accumulated enough critical mass inside the customer. We've accumulated multiple usually different sort of areas of support inside the customer that now add up to this number. More conceptually what we're aiming for and what we believe is how the database market is going to keep evolving. Because to your point, customers want to limit the number of databases that they're running, not expanding. There's no -- the operational burden of having 15 different databases for an enterprise customer is significant. And there's not that much benefit. So we believe that number over time will be coming down, and we want to be one of those standards, if you will. And so how do you become a standard? You become a standard through demonstrating strong performance and great customer service and report across multiple workloads. And then at one point, it is either implicitly or explicitly known that a particular kind of application or an application meeting a modern database or any application should go to Mongo unless there's a really, really, really strong reason. And look, we're very early in that journey. We have some customers where we've reached that status. But the reality is, is that as evidenced by our market share, that it's just the very beginning, and that's something that we have to work towards and look forward to in the future. If you circle back to Q4, I will say there's 2 things to keep in mind. The first one is generally our biggest new business quarter. And so that's a -- that's true for EA and Atlas, but obviously, EA has more of this sort of upfront revenue recognition impact. So that's a thing to keep in mind. In last Q4, we called out, it was a particularly strong EA quarter, and we called out one, only one large customer. There was a $3.5 million benefit from a multiyear EA contract, but it's not the only one. It was just generally a very, very strong EA quarter. So now we'll be comping against that, right? And that's just something that we are cognizant of and want to make sure we remind investors of as we're sort of approaching the end of the year here. As far as sort of looking forward when it comes to pipeline generation, we see great engagement with a customer. This is what Michael was talking about. We had good Q2 in terms of new business, we had obviously a record quarter when it comes to customer additions. Secular trends are only accelerating. All that, at least to very good engagement. But juxtapose to it is obviously the macro uncertainty, which isn't going away anytime soon. And we'd be shortsighted to assume that just because we had a very strong Q2 and exceeded our own expectations that we sort of checked that box and figured it out. So we have no doubt that they will continue playing on sort of how we perform and then particularly in Q4 because of the big part of the new business impact. The other thing and goes back to the other part of the conversation is, we mentioned that we grew slower in existing customers in Q1, and grew better than in -- in Q2, we grew better than Q1, but still slower than history. So if you think about...

So is this like Atlas customer expansion?

That's right. So if you think about all our customers in Atlas that we had at the beginning of last year, they grew slower in Q1, better but still slower in Q2, and we're projecting slower growth for the rest of the year as the macroeconomic uncertainty continues. So you have this sort of -- by the time you get to the fourth quarter, you have a year-over-year impact of 4 quarters of slower than historic growth. And that ends up impacting the year-over-year growth rate. That's not an EA phenomenon, that's an Atlas phenomenon, but it also fits into how you think about the guidance and the outlook for the rest of the year. And Q4, in particular.

Maybe there's 2 other things that I'd add, just so people can understand. So the first is an accounting impact that people just need to make sure they understand this. So Serge mentioned the strong Q4 impact or new business from Enterprise Advanced last quarter -- last year -- Q4 of last year and how that presents a tough compare. You call that the one large customer that we called out, that was $3.5 million of sort of incremental revenue in Q4 as a result of the multiyear nature of what they did. And just so people understand what that is, is it's not just that you need to sort of subtract it from the denominator, but you have to acknowledge that you're sort of missing it from the numerator. And by that, I mean, just so people are really clear on how the accounting works on Enterprise Advanced is we've said about roughly 25% of the annual subscription is in the upfront term license revenue. And so the way that would work is if we had $100 subscription from a customer, at the time that silicon starts, we have recognized $25 for term license revenue and then the remaining $75 would be recognized over the 12 months of that first year. And then I get to year 2. And so in month 13, I have another $25 and then up to $75 for the second year, we're getting recognized over the remaining 12 months on 13 to 24, sort of et cetera, et cetera. For example, you did a 3-year deal with the customer, right, same setup, right, 3 years, $100 a year, $300, you recognize $75 upfront, right, for the term license revenue for all 3 years. And the remaining $225 will be spread over the remaining 36 months of the contract. But what that means is when you get to month 13, not only did the base period have this sort of inflated $75, but the numerator doesn't have the $25 because you've already recognized it, right? And so that's when we talk about this increased variability and the reduced comparability that's caused by 606 accounting that requires to do it this way, that's sort of what we mean. So hopefully that's helpful for people to sort of understand. And then the second thing that I would just say, on the broader question -- on the broader topic, is of your $1 million large customer, how you want to phrase the question is for -- there are a small number of those customers where we have the majority/entirety of their deals then, but most of the customers that are our largest customers, whether you say customers over $1 million, top 50 like how do you slice and dice it, we have single-digit wallet share of what they spend on databases, which just sort of underscores the fact that it's such a large market and most of them really early on in our penetration of it.

Got it. And Michael, when you talk about the penetration, are you -- is this kind of -- how do you define that broader market? Is that both kind of like analytics and operational databases? Or is this kind of like more in the core type of database functionality that you often...