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

Welcome to our next session. I'm really happy to have the team from MongoDB on.

Michael, you just had a very strong Q3 results last night. Maybe from your perspective, like what stood out most like? What was the -- what were the highlights from your perspective?

Sure. We are happy to go and do it. And first of all, just thanks for having us. Good to see you Raimo. And -- yes, so we reported last night, just quickly for those who don't know revenue just under $227 million, up 50% year-over-year. Subscription revenue was up 51% year-over-year. Atlas is now 58% of the business and grew 84% year-over-year. So really just another strong quarter of new additions, plus 2,000 new customers, plus 300 on the direct sales side. So just a really strong quarter overall. Strong in terms of new business. Really broad-based strength across industries, across geographies. In particular, we saw not just great customer additions, but real strength in the Atlas expansion in the cohorts. I think that was great. But it was also a really strong EA quarter. So I think a lot of times, people sort of assume that Atlas is -- always steals the show and certainly given those growth and its market reception, I clearly understand why the focus is there. But to me, when you look at the combination of the two, it really just sort of speaks to the size of the market that we're going after, $74 billion by IDC standards today going to $121 billion in 2025, the strength of the product market fit. And then all of that matched with really good execution in the marketplace. So I think our go-to-market teams and everything else are really executing well. And so we're really pleased with the quarter.

Yes. Okay. So you took half my questions now. That's not good.

No, no, no... We can go in all the depths.

The -- if you think that Atlas' growth really stood out to me, the 84%, can you talk a little bit about some of the factors? Because it did accelerate again -- So from that perspective, like I wanted to dig a little bit deeper there in terms of what's driving that?

Yes, absolutely. So both Atlas and Enterprise Advanced exceeded our expectations in the quarter, but let's talk about Atlas specifically. Especially if you're looking at -- and I know you all slice and dice things in a million different ways. But if we look at things sort of in two different ways, sort of the year-over-year basis and then the sequential basis, we'll try and kind of call out a couple of factors. I think on a year-over-year basis, it was a tougher compare, and we have a tougher compare in Q4 as well, given that in the year ago period, we have sort of seen that bounce back from that sort of COVID slowdown that we've seen in Q2. So really strong performance. And then sequentially, when you think about it, what's really interesting is while we're adding new workloads and adding new customers, an Atlas customer tends to start relatively small and then builds unless it's sort of a high-end enterprise workload. But what we see over time is that the new business that you win in any given quarter doesn't usually materially affect the revenue in the quarter for Atlas. Again, Atlas is consumption-oriented, right? So it's all based on the usage. And so usually, when you're looking at the sequential basis, what covers the performance relative to the prior quarter tends to be how do the existing apps, the existing workloads that you have, how are those performing? We have a fairly -- we're building up some history here. We're only a few years into it. We're building up some history. And there's a pretty normal kind of healthy range that those applications will expand that. And in Q3, we saw those expand towards the higher end of that range, which is great and wonderful to see. But that's sort of in addition to the new applications and the general overall success that we're having really is what drove the strong Q3.

Yes, yes, yes. And then now how do you think about Atlas more longer term? So it's now 58% of total revenue, which means that like the cloud has taken over. Is there a level where that might level out, because you have guys like Barclays, et cetera, highly regulated industries that always will be, or will be tough to kind of bring over? So is there a level or where do you see that number going?

Yes. So over time, certainly, we'll continue to increase as our current expectation. But you're right, not everyone is ready to adopt the cloud now or even over the next couple of years. I know a lot of times in conferences like this and in other settings, there's a lot of talk of like cloud, cloud, cloud. But I think people missed the fact that there are a lot of industries and a lot of companies that really aren't ready yet to adopt public cloud. And that's part of what continues to drive the success of Enterprise Advanced. So the way that we continue to think about it is we really run the business on a channel basis, as -- we provide a lot of disclosure from a product level standpoint, but we really run the business on a channel basis. The self-service channel is basically all Atlas. The mid-market team is almost all Atlas at this point. Those tend to be early technology adopters, more cloud-forward companies, things like that. And so ultimately, where the mix shift will go will depend on how quickly do enterprises adopt Atlas. And what we've seen, and as we've talked about and as we've highlighted with some of the vignettes and everything else, is increasingly we're seeing lots of enterprise workloads. Earlier in the year, we called out that 60% of our customers over 100,000 would have qualified over that threshold just based on their Atlas spend. And we're seeing larger and more meaningful workloads that are running on Atlas, and we'll talk a little bit about that. I'm sure we'll touch on some of the renewals discussion. But some of those are large enterprise meaningful mission-critical workloads that are running on Atlas. And so as more and more enterprises are comfortable adopting that, we'll see that shift. And so I think the ultimate percentage really depends on both your time frame, like how long a horizon you will look out and what do people like Barclays and other people, who are more regulated industries and everything else do.

Yes, yes. Okay. And then the other question on Atlas was like, remember, if you think about the evolution of Atlas, there was a lot of self-service, then your products kind of matured and direct took over and now direct is kind of the main motion. And you and I always had that debate about like how do I have to think about self-service and the growth there versus a direct because self-service is feeding into direct. But like what are you seeing there in terms of that growth on both sides? And I'm asking because this quarter or yesterday, we did see self-service pick up a little bit again in terms of extra growth. So it does feel like both motions can work.

Yes, absolutely. Both motions definitely can work. And I think that we're getting increasingly better over time at how to benefit from the two, not just think about one and the other, each in isolation. And so we're both continuing to look at the self-service channel as a great, efficient customer acquisition channel for a mix of business. And then for the workloads that wind up being having more potential, it's effectively almost like a lead gen channel, if you will, for where it makes sense to interject a sales relationship. And I think for us, we've been continuing to iterate on and experiment around or what are the right signals, what are the right moments, what are the right actions that we should be taking to optimize the outcome and the customer journey and ultimately how much money that they're spending with us. So for example, there are some customers that will naturally grow on their own, don't really need or won't benefit from a sales interaction, in which case, we shouldn't move them from the self-serve channel to the sales sold if we're not going to positively inflect their arc because all you're doing is paying a salesperson to go do something that you would have gotten anyway, right? So we try to be really disciplined and intelligent about what are the incremental outcomes from the investments that we're making and making sure that we are increasingly finding or focusing the efforts of the things that really make a difference. And so I think in the big picture, they're both very viable paths to customer acquisition, self-service and direct-to-customer acquisition and increasingly, we're getting more efficient about how to use the two together.

Yes. Yes. Okay. Got it. And then do you...

Actually, one other thing that I will say, though, is we report lots of different slicing and dicing of the data that I suppose -- I do think it's important for people to understand that there is a little bit of a natural cap or ceiling or whatever you want to call it, on the self-serve business because the largest customers almost always will benefit from a direct sales relationship with us. And so you will see those move through the sales sold channel. And so if you think about Atlas overall, Atlas overall is growing at 84%. Though it is a very good quarter in the self-service channel, but if you look at it on a year-over-year basis, the self-service channel grows sort of something like a 26% kind of mid-20s number is what you're going to calculate. And that's not because the business isn't doing well or we're not good at identifying self-service customers, it's just there is the sort of cherry picking, if you will, [ where the largest best most meaningful dollars ], small in terms of percent of customers, but disproportionate in terms of dollars, are the best candidates to have a direct relationship with, and we can further accelerate their growth from there. And I think indiscernible] understand that dynamic.

Yes, yes; yes. So -- and do you have like a business leader? -- sorry, is there a business guy that self-serve [indiscernible] further, is it like all one? Like I feel sorry for that guy because his best company is always handing over?

Well, yes, so there's a growth marketing team that's focused on self-serve and driving those customers, and they understand the dual purpose that channel serves. So there aren't any hard feelings.

Yes, yes. Okay. And then the customer acquisition was strong yesterday if I look for the model again. And if I look -- I mean, as an organization, you're adding like if I hope I'm not wrong here, but about 2,000 new customers a quarter. That's a damn big number. Like can you talk a little bit about like how you drive that? And how sustainable is that? -- I mean it sets you up very well for the future, but like it's a crazy big number as well to kind of achieve every quarter.

Yes. So a few thoughts. We've been at that level pretty consistently for the last couple of quarters. I think you're right. I think it mostly speaks to the size of the market and the strength of the product market fit. And we're pretty consistent in the last few quarters, seen about 300 net additions in the direct sales channel. And so most of that -- 2,000 would be on that self-service side that we're describing. And I think it really just speaks to the popularity of MongoDB. And if you take a step back and you think about the tens of millions of downloads of MongoDB and the developer popularity of MongoDB and the fact that MongoDB has the database that developers most want to use and all of those kind of dynamics. That's really what we're tapping into and we're getting sort of more efficient around it. And certainly, the number will fluctuate quarter-to-quarter. But in general, it feels like we're still pretty early on in tapping that opportunity.

Yes. Okay. And then the other -- I think that's the last question on Atlas now because then you kind of shut me up as. The one thing that was interesting, and it's funny without deferred, we would have never known it, you had some very big early renewals in Atlas. And you've got several questions including me on the call is like what's going on there that seems not normal? Can you talk about? I mean, again, it shows the strength in the business and how excited people are to work with you, but like talk to that a little bit more because that was unusual this quarter?

Yes. I think it also underscores the mission criticality of Atlas. But Serge, do you want to cover that?

Yes. Let me jump through this and then Michael feel free to follow up. So we had a handful of our largest Atlas customers do early renewals of their contracts in Q3. Some of those were contracts that would have come up for renewal in Q4 and some were actually from Q1. And some were multiyear, some were single year contracts. But what they all did was -- these were customers that were consuming in advance of their commitment. So their run rate was higher than what they already committed to us. And they had incremental visibility and confidence that they will spend even more either because of the growth of our existing apps or more likely because they knew they will be bringing more apps to the platform going back to the whole sort of journey to standardization that Dave has talked about last night. And so they approached us to do bigger deals, bigger commits. And those tend to be built annually in advance. So those have impacted our deferred revenue when those contracts were signed. Importantly, though, this has nothing to do with our revenue. So Atlas revenue is recognized on consumption. So it's not like an EA where an early renewal would generate the 606 license recognition, and then we would need to talk about it in the context of revenue in the quarter. This was no impact. We just wanted to call it out because it was unusual jump in the deferred and wanted investors to understand the mechanics on the balance sheet, but also to Michael's point, what this really speaks is sort of the -- where we're heading with Atlas in terms of increased sort of strategic importance of it and the willingness of some of our largest and most committed customers to commit even more.

Yes. Okay. Makes sense. And then the -- I'm smiling, now while I'm asking it, but like what's interesting is you can't hold us not to look at billings because it's not a good indicator. But all of a sudden, if I have like things like this, they show up in billing, like it's not shown up on revenue, I get that. But it did show up in billings, like, its there like maybe an argument to say maybe we should start paying more attention there again? Or was it just like, this quarter was like so unusual that you're not going to see it for a while?

I would definitely not suggest that this is a reason to you turn and go back to billings. It's a -- we'll see more of these contracts in terms of the strategic importance, but what was really important about this quarter that it was unique in terms of its sizing and it was literally some of the largest Atlas customers who renewed early, so I would view this definitely as an [ outlet ].

Yes.

I would also add, Raimo, maybe just like -- and this maybe gets too philosophical, but hopefully, it's helpful, is the way that I think about it if I were an investor is that in sort of an [ annual recurring ] software business, the reason why I think of deferred revenue as interesting or why I like a calculated billings number is I feel like it maybe helps me get a peek ahead, right, as to where things are going. And I think in this example and in MongoDB, as we talked about like even earlier this year, we talked about how something like 40% of ARR doesn't even pass-through deferred, right, given Atlas and some of the usage base in self-service monthly billing arrears and all those kinds of things. And I think in this example, even though it's a big number from a revenue standpoint, as Serge walked through, those customers were already spending at certain run rates, right? And those run rates were usually in excess of their existing commitments. And so that's already all in the revenue, right? And so when you think about it, it's not that hopeful of a forward predictor. I think all it really signals is it is helpful for some people is to sort of underscore like the mission criticality of Atlas, and it clearly indicates that people have increased confidence to -- that they're going to grow even further, adding new apps and everything else, new workloads to Atlas, which is sort of a good sign. But from like a quantitative standpoint, it's sort of a hard number to get any signal relative to the noise.

Yes, yes, yes. Okay. Perfect. And then -- I mean, is there -- if you get those bigger commitments like -- it's nice for you -- do you give them like a big economic incentive to commit more? When I talk out smoothly, they kind of want to have that, but it doesn't sound like for you, it really is more a function of the customer wanting it?

Yes. I think it's more of the customer wanting it. The way that I think about it is Atlas in general is pretty sticky and -- as is MongoDB overall. When someone is committing to spend in excess of their run rate, though, a commitment is worth something. And especially if it's a forward commitment, right, because now it starts to align the incentives where we both have an incentive to go find more workloads, right, to help further drive the consumption. And so that's great. So we like that. But we've been really clear, first with our sales force and the sales force, either educating or reminding customers that if you're used to a certain discounting approach within software, where as you buy in much bigger qualities, you start to see some pretty meaningful discounts because the marginal cost of the software is very limited. I think people understand that when you're delivering products as a service, particularly database product where there's the underlying storage and compute is part of that. That the -- yes, you can pick up some incremental discounting, but it looks nothing like typical kind of software discounting. That said, if you've got workloads, you know they're growing. You know you're going to add more workloads. And you've got confidence in our partner [indiscernible], like even if I can optimize it or eke out a little bit of incremental discounting, I'd rather do that than not. And that's kind of how the dialogue is. But it usually has to be customer initiated.

Yes, yes, yes. Okay. Yes, makes sense. And then talk a little bit about like late shift gear a little bit to EA, that is, Enterprise Advanced and in the -- was kind of better than I would have expected. Can you talk a little bit about what's going on there? Is this like a post-pandemic world and people are [ trying to expanding again ]? Or like how do I have to think about this?

Yes. So I think -- I don't think there's any major narrative theme, pandemic or otherwise, that I would just go to other than the fact that not everyone is ready for the cloud in any given quarter or for any given sales rep, the mix of accounts, the mix of transactions will govern, particularly within enterprise -- the enterprise channel, will sort of go the mix of Atlas versus EA. I think what it underscores is the value prop and the value of having multiple different ways to consume and to meet the customer wherever they are in their cloud journey. And increasingly, we are talking about, if not yet seeing, people who are using and adopting Enterprise Advanced thinking about the cloud and/or at least understanding that using Enterprise Advanced will ease their future cloud transition, right, because they can just move to Atlas, and it's much easier than having to rewrite or replatform the application. And there are certain industries, for sure, that tend to be more or maybe less in the cloud, financial services, health care, like things like that tend to still be a little bit more self-managed on-prem. But I don't think there's any -- Serge, you can jump in or maybe disagree, but I don't think -- I wouldn't add some sort of macroeconomic supply chain pandemic theme or anything else like to that. I think it's more just sort of a mix of deals in the quarter and where are our customers cloud wise.

Yes. Okay. And then the -- if you think about it, like we -- the [ compression ] came up and -- or you kind of mentioned on the conference call, it's kind of more workload. So is EA a little bit like -- could I think about it as like there are like guys like Barclays that are kind of on-premise and they're expanding usage? So it's almost like a similar message as an Atlas is just that its a different customer with different kind of situations for the customer, but it just shows like there's underlying strength and desire to do more Mongo, and that's why you see those numbers?

Yes, directionally, I think that's right. Serge, I don't know if you want to elaborate or I think...

I think that's exactly right. I think that -- and Barclays is a great example. It's an example of a company where our market share inside the company is still relatively low, and there's great opportunity to pick up incremental workloads and the rep dedicated to that account is highly economically incentivized to make that happen. And so we are [ indifferent to ] which technology you are consuming. We know they were grossly underpenetrated in the market, particularly relative to the quality of the product and the product market fit. So we see opportunities sort of on both sides and what drives which one you adopt is your IT strategy as opposed to sort of anything that we can dictate or want to drive.

Yes. Okay. Let's shift here again a little bit. Though our net expansion you talked about was very strong, talk a little bit -- I mean, I guess, like, look, we had a lot of upsell on Atlas, EA was doing really well. What's the -- like you always give us like that 120 number? Like are you kind of how do we have to think about that motion? Because it does feel like you're kind of decently above that and at some point of argument even to change that 120 number. I don't -- I'm dreaming now. But like...

Yes. So maybe just taking you a step back. As you think about our net expansion, it comes into two flavors. I think that's important because it might be different than some of the other companies that you're discussing. One is obviously the growth in existing workloads. And that's important and helpful. But overall, with some notable exceptions aside, like a gaming company or other similar sort of highly sort of elastic workloads, we're closing to grow relatively slowly, particularly after the first couple of years. So really, the name of the game in our net expansion is incremental workloads. That's true in EA. That's true in Atlas. And that's sort of this constant desire and opportunity to keep spreading inside of our existing customers find incremental workloads and eventually sort of move our way towards the standard. So I just want to call that out because sometimes we get questions that sort of seem to imply that 120 plus is what we get when we show up on the first of the year, and that's not at all true like we definitely have to work for it. Honestly, the decision to disclose 120 plus was made sort of prior to the IPO because we went around and asked people what is best-in-class and some people said 115, some people said 120. So we said, fine, we'll just say 120 plus, obviously, for as long as that's true. We think that would fundamentally sort of underscores the strength of the platform and the opportunity ahead of us. And the reason why we're disclosing it that way is because we thought that if the number went from like 124 to 128 or the other way around, they may generate more questions than is necessarily sort of helpful or valid because ultimately, it's a very strong number. And our goal is just to keep it above that number for as long as possible.

Yes. Okay. Okay, makes sense. I mean, you've got lucky anyway, because like I was talking with some other software vendors and they kind of were not as lucky; they have to give like specific there.. [indiscernible] The -- just changing gears in the last few minutes. Like as you go into Q4, like a lot of the other companies started to point out more COVID, new strains, et cetera, that's coming into play that kind of make them a little bit nervous and got the guidance may be a little bit more conservative. What are you seeing out there? Is that something that you're concerned about like how are you thinking about it?

Yes, I guess I just -- I'd say a few things. There's an enormous amount of uncertainty. We've clearly seen that. And whether it's some new variant or anything else, certainly, we all have been dealing with this for the last several quarters. I think there's nothing specific that we can sort of point to. And if I think about just sort of our own evolution, in March of 2020, we were one of the first or only or whatever, like specifically talk about COVID and attempt to quantify it and everything else as we've sort of gone through this time period. I think the thing that we've come to believe and have increased conviction of over these last several quarters is despite the uncertainty, I think we've been able to execute well in this environment. Now that's certainly not a given. There's certainly risks and all that kind of stuff. But part of our optimism, as you can see in the Q4 guide and raising the full year by more than the beat and everything else is a confidence that despite the fact that things may be choppy in that there may be uncertainty, is we now, like I said, have sort of several quarters of executing well against that. And so rather than just applying some sort of brush of conservatism to it that, hey, there's uncertainty, we've sort of said, look, we now have some data that shows that even amidst the uncertainty and the volatility, we've been able to execute and manage through and we don't have any reason to believe that we can't do that. And so that's what's contributed to the strong guide and increased confidence and optimism.

Yes. Okay. Perfect. And last question for me was on costs. So obviously travel was slightly less than you planned, and some of that is coming back before but not fully. You had some savings on the marketing side. Can you just speak to that in the last couple of minutes?

Yes, Serge, you want to talk to that one?

Yes. So let's start with travel expenses. So our -- at the beginning of the year, at the beginning of Fiscal Year '22, when we provided the original operating loss guidance for the year. We said that we expected to see incremental costs of $20 million to $25 million in Fiscal Year '22 compared to Fiscal Year '21 related to travel event and workplace type expenses because of what we at the time thought would be normalization, sort of a post-COVID normalization in the second half of the year. Now we didn't assume that it would normalize fully. But we did expect that things would get far closer to pre-COVID levels in the second half of the year because when we're providing guidance, that was like the first round of vaccines and people were optimistic. Obviously, since then, will this become more complicated? We've seen a bit of an increase in those expenses but not as much as we had expected. So whereas at the beginning of the year, we were thinking to $20 million to $25 million, now in our latest guide, that view is $9 million to $12 million and that's, again, like back half loaded. So much of that is coming in the second half. So our construct around these expenses, frankly, is that -- this is a onetime saving. We do expect the world to normalize whether it's 100% to pre-COVID levels or some lower number. And is it for full year as we go into it or thinking about next year? Obviously, there's a lot of uncertainty, and we'll have to sort of contend with that as we continue with our planning for next fiscal year. But we are committed to our real estate strategy. We think events are a great marketing sort of tool that is important for our brand building and travel is important internal and external for a variety of reasons. So we fully expect those costs to normalize as the environment permits it to. But we are sort of benefiting from this onetime, calling it saving is wrong, but just we didn't rebound as much in the second half. But as we think about sort of planning for next year, we're sort of planning for our continued investment in the business and then the post-COVID bounce in expenses will sort of be what it is.

So this wasn't less sales hire or something. This is really like more travel and stuff like that.

So T&E, in-person events, office space, stuff like that.

Yes. Okay. Makes sense.

[ We have just to keep it. ] To give you an example, we have less attendance than we expected in our offices. Therefore, we need less snacks, less meals like everything...

People eat more that are out there. Yes.

And then -- so that was one thing we called out. The other thing is more tactical in the very near term, that we thought it was important for people to understand to put the strong operating loss performance in Q3 in context. Our CMO, Peder Ulander, started at the beginning of Q3. And he did what probably you and I would do if we were in their positions, which is he wanted to review what the team was doing, he sort of put certain larger ticket items when it comes to program spend on hold until he sort of went through them, the review is complete. He's mostly left, things have changed, he sort of reallocated certain budgets here and there, but that benefited Q3 and we're going to be back to normalized marketing spend in Q4.

Perfect. Okay. Great summary. Congrats, this was a very strong quarter all across. So amazing. Nice to be part of that first as an analyst. But hey, thanks for joining us again, and all the best for Q4 next year when the -- [ hope to see you soon ] hopefully.

Yes. Thank you so much.

Yes. Thanks, everyone. Thanks for everything. And thanks for having us.

Yes, of course.

Take care.