Good day. Welcome to the MongoDB second quarter fiscal 2021 earnings conference call. All participants will be in a listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. Please note that this event is being recorded. I would now like to turn the conference over to Brian Denyeau with ICR. Please go ahead, sir.
Great. Thank you, Chuck. Good afternoon, and thank you for joining us today to review MongoDB's second quarter fiscal year 2021 financial results, which we announced in our press release issued after the close of the market today. Joining the call today are Dev Ittycheria, President and CEO of MongoDB, and Michael Gordon, MongoDB COO and CFO. During this call, we will make forward-looking statements, including statements related to our market opportunity and future growth, our financial guidance, and the anticipated impact of the COVID-19 pandemic on our business and the results of operations.
For a discussion of the material risks and uncertainties that could affect our actual results, please refer to the risks described in our SEC filings, including our most recent annual report on Form 10-K and quarterly report on Form 10-Q. Any forward-looking statements made in this call reflect our views only as of today, and we undertake no obligation to update them. Additionally, non-GAAP financial measures will be discussed in this conference call. Please refer to the tables in our earnings release on the investor relations portion of our website for a reconciliation of these measures to the most directly comparable GAAP financial measure. With that, I'd like to turn the call over to Dev.
Thank you, Brian, and thank you to everyone for joining us today. I will start by reviewing our second quarter results before giving you a product and go-to-market update. Looking quickly at our second quarter financial results, we generated revenue of $138.3 million, a 39% year-over-year increase and above the high end of our guidance. We grew subscription revenue 41% year-over-year. Atlas revenue grew 66% year-over-year and now represents 44% of revenue. We had another strong quarter of customer growth, ending the quarter with over 20,200 customers. These would be strong results in any environment but are particularly notable during a global health and economic crisis. I'm especially proud of our employees across all facets of the business, have adapted to working in a remote environment and execute at a high level.
Their passion and commitment to our customers and to the business in this difficult time have been outstanding. Over the past five years, we have conducted detailed employee engagement surveys semi-annually to understand how our employees feel about the company. Tellingly, in our most recent survey conducted this past July, the scores on employee engagement and on their confidence in the company's future prospects were the highest ever recorded. Every business is quickly becoming a software business, which means the value proposition of any business is increasingly enabled, delivered, or defined by software. Consequently, a key unit of productivity of a business becomes the developer, as developers create enhanced applications that lead to a competitive advantage. Since our inception 13 years ago, we put the developer at the center of our universe, and we introduced a new class of database based on a document model or JSON.
JSON is a data format that's easy for developers to read and write and machines to parse. We have always believed that JSON is the best way to work with data. As tech-savvy companies around the world adopted the document model, the industry took note. First Microsoft, then AWS, and most recently Oracle tried to emulate what we have done. Oracle recently made public statements that acknowledge that JSON has become the main data model for new applications and that developers love JSON because it provides support for dynamic schemas, consequently making it easy to make changes. I believe it's noteworthy that the industry now agrees with our fundamental premise. The document model is simply the best way to work with data. Our mission continues to make it stunningly easy for software developers to work with data, wherever it resides, to drive innovation and create value.
At MongoDB.live, our user conference held in June, we made several significant product announcements that further advance our mission. With the introduction of MongoDB 4.4, we delivered a number of additional feature enhancements that push the envelope of what it means to be a modern database. With new capabilities in our query language layered on the most flexible distributed systems architecture anywhere, developers can build the most sophisticated transactional and analytical applications securely at scale. MongoDB Atlas Data Lake is now generally available and allows teams to query and analyze the structured and unstructured data in their S3 buckets using the MongoDB Query Language. Atlas Data Lake also supports federated queries, which means teams can submit a single query and analyze operational data in Atlas alongside their data in S3.
We announced the beta of Atlas Online Archive, which completely changes the economics of large data sets by allowing users to define rules that automatically archive data from their Atlas database to low-cost cloud object storage. Best of all, customers retain the ability to seamlessly query their archived data with no extra effort. The general availability of Atlas Search allows developers to deliver rich search experiences on top of their data in the cloud without needing to deploy, learn, and manage a separate search technology. Atlas Search uses the MongoDB Query Language and is fully managed. Finally, we unveiled MongoDB Realm. It combines the popular Realm Mobile Database we acquired last year and the serverless data access, data movement, and data manipulation services formerly known as MongoDB Stitch.
Our core component of MongoDB Realm is Realm Sync, which is available in public beta. This edge-to-cloud data synchronization service between Realm Mobile Database on the front end and MongoDB Atlas on the back end solves one of the most challenging data problems for mobile developers. Our recent product announcements enable customers to use MongoDB for a broad set of use cases and represent a significant step forward on our journey to deliver the preeminent modern data platform for the developer. Our strong second quarter results reflect the fact that our go-to-market organization executed exceptionally well under challenging circumstances. It is in difficult times when world-class go-to-market organizations separate themselves from the mediocre ones, and we are proud of our team and our sophisticated approach to growing our business.
Our second quarter results also demonstrate the increasing efficiency of our go-to-market efforts as we are driving increased synergy across our four sales channels, our field sales force, the inside sales team, our self-service channel, and our partner organization. Starting with the self-service channel, where over the past two years we assembled a strong growth marketing team, built out processes and the infrastructure to scale our self-service business, and implemented a way to rapidly experiment and launch programs to increase the size and quality of our pipeline. We're pleased by the traction we're seeing, particularly in terms of acquiring new Atlas customers, which have achieved a third consecutive quarter of record growth. As our sophistication managing the self-serve business grows, we're finding ways to accelerate the growth of the overall Atlas business.
We do this by identifying self-serve customers based on product usage signals that benefit from direct engagement with our sales organization. With more attention, service, and support, we see a significant acceleration in customer spend. As we have become better at identifying the self-service customers with high-growth prospects, it has given our sales teams a more efficient way to prospect our self-service customer base. The end result is a self-serve channel that generates significant value on multiple levels as an important revenue generator in its own right and as a source of excellent leads for inside and field sales teams. Turning to our inside and field sales teams, over the past year, we have also gained conviction that new Atlas customers, irrespective of their size, when given proper resources and support at the outset, grow rapidly.
This is true not only for customers who transition from self-serve, but also for customers launching brand new applications on Atlas or migrating their self-managed instances of Community Server onto our platform. As a result, we have adjusted our approach to increase the velocity of acquiring new customers. This year, prior to the outbreak of COVID-19, we changed the incentives of our sales team to focus more on landing new customers onto our platform and less on size of the initial commitment. Moreover, we are using our professional services organization strategically to set up customers for success early to give them more confidence in using MongoDB. The result is a third consecutive quarter of record customer growth with notable strength in our inside sales channel. Lastly, the partner organization is sourcing or influencing a growing percentage of our business, especially with cloud providers.
We have seen strong momentum on deal activity with all three major cloud providers, and each cloud provider is providing more marketing dollars, adding more incentives for their salespeople to work closely with MongoDB, and dedicating more headcount to better support and grow the partnership. Given the fundamental advantages of the document model and the breadth of our product offering, customers recognize that MongoDB is the clear choice for their digital transformation plans. The combination of our leading modern data platform and our sophisticated go-to-market efforts allow us to deliver consistently strong performance. Now, I'd like to spend a few minutes reviewing some of the customer wins and interesting use cases from the second quarter. One of the world's largest telecommunications providers is currently developing a next-generation location service that can pinpoint mobile locations with centimeter-level accuracy.
Their need for a flexible data model with rich indexing and translytical capabilities and a data platform that can scale quickly led them to choose Atlas this quarter. Adeo Group, the world's third-largest holding company in consumer goods for do-it-yourself and decoration with an international network of 14 brands across 13 countries, moved to Atlas in order to standardize app development among their developers and accelerate time to market as e-commerce demands grew during COVID-19. Agero, a leader in the digitalization of driver assistance services, serving more than two-thirds of insurance companies in the U.S., chose MongoDB to help power its next generation of software-enabled driver safety services and technology. MongoDB Atlas has allowed them to move from one monolithic application to microservices and provide a better digital experience for the 115 million drivers they protect each year.
Highspot, the sales enablement platform that makes every conversation count, helps customers worldwide empower their remote sales efforts, which is now more crucial than ever because of COVID-19. They recently standardized on MongoDB Atlas to drive strategic growth and improve customer satisfaction. Symphony, the cloud messaging and collaboration platform for the financial services industry, chose MongoDB Atlas to support the evolution of their architecture and foster efficiencies. Atlas will allow them to adhere to crucial security policies and provide the best and most reliable experience to their customers at scale. Lastly, one of the world's largest cloud providers has built a security solution on MongoDB that provides rich visibility, control over data traffic, and sophisticated analytics to identify and combat cyber threats across their cloud services.
This cloud provider has increased its commitment to MongoDB, making us the primary technology for handling immense workloads in a highly flexible yet secure environment. In summary, we are very pleased with our performance in Q2. While the macroeconomic environment remains challenging, we are executing well and remain focused on the enormous opportunity that lies ahead. We continue investing to capture this opportunity and, in particular, recruiting world-class talent to help us maximize our potential. To that end, in the past weeks, we welcomed several important additions to our Senior Management Team. Mark Porter joined us as our Chief Technology Officer in July. Most recently, CTO of Core Technologies and Transport at Grab, Mark is a 30-year veteran of the database industry. He was an early member of the Oracle Database Kernel group, and later was a general manager of the AWS RDS business.
Mark originally joined MongoDB as a Board Member earlier this year, but he got excited about the opportunity in front of MongoDB and wanted to be directly involved. Very few people understand the relational database technology and its limitations better than Mark, so we're excited that he voted with his feet and joined the world's most popular modern database company in a senior leadership role. In early August, we welcomed Harsha Jalihal as our new Chief People Officer. Harsha is a seasoned HR Executive who has been at the forefront of scaling large tech companies as well as managing complex enterprises. Most recently, Harsha was the Vice President of HR at Unilever and was responsible for delivering end-to-end HR strategy and operations for the U.S. business. Prior to that, she was an HR executive at Cognizant during its hypergrowth phase.
Finally, in mid-August, Rishi Dave joined us as Chief Marketing Officer. Rishi is a veteran technology marketing executive with more than 2 decades of experience, most recently at Vonage, where he was the company's Chief Marketing Officer. Prior to that, Rishi was CMO at Dun & Bradstreet, a cloud data and analytics provider, and earlier had a senior leadership role driving Dell's digital marketing strategy. Harsha, Mark, and Rishi join a talented and dedicated senior management team that is performing at a high level, as our Q2 results clearly demonstrate. We are excited for what we can all accomplish together as we continue disrupting one of the largest markets in software. With that, I'll turn it over now to Michael.
Thanks, Dev. As mentioned, we delivered another strong performance in the second quarter, both financially and operationally. I'll begin with a detailed review of our second quarter results and then finish with our outlook for the third quarter and full fiscal year 2021. First, I'll start with our second quarter results. Total revenue in the quarter was $138.3 million, up 39% year-over-year. Subscription revenue was $132.5 million, up 41% year-over-year, and professional services revenue was $5.8 million, up 11% year-over-year. To put our performance in the quarter into perspective, we thought it would be helpful to provide an update on how COVID-19 has impacted the growth of our business. First, let's talk about new business. The negative impact of COVID-19 on new business in Q2 was less than we had expected.
As Dev explained, our various go-to-market teams had executed well in the 1st fully remote quarter since the pandemic started. 2nd, in Q1, we noted that we observed a slowdown in the growth in spending from our existing Atlas customers, particularly in the self-serve channel. In mid Q2, as the economies around the world started the gradual process of reopening, we saw an improvement in the rate of growth of our existing Atlas customers. While the trend still remains below historic levels, the improvement we experienced in Q2 increases our confidence that the slower-than-historical growth is simply a macro phenomenon. Overall, Atlas' strong performance continued to be the biggest contributor to our growth. Atlas grew over 66% in the quarter and now represents 44% of total revenue, compared to 37% in the 2nd quarter of fiscal 2020 and 42% last quarter.
During the second quarter, we grew our customer base by over 1,800 customers sequentially, bringing our total customer count to over 20,200, which is up from over 15,000 in the year-ago period. Of our total customer count, over 2,500 are direct sales customers, which compares to over 1,850 in the year-ago period. The growth in our total customer count is being driven in large part by Atlas, which had over 18,800 customers at the end of the quarter, compared to over 13,200 in the year-ago period. It's important to keep in mind that the growth in our Atlas customer count reflects new customers to MongoDB, in addition to existing EA customers adding incremental Atlas workloads. We had another quarter with our net ARR expansion rate above 120%, despite the impact of the macroeconomic environment.
We ended the quarter with 118 customers with at least $100,000 in ARR and annualized MRR, which is up from 622 in the year-ago period. Moving down the P&L, I'll be discussing our results on a non-GAAP basis, unless otherwise noted. Gross profit in the second quarter was $99.7 million, representing a gross margin of 72%, compared to 73% last quarter and 72% in the year-ago period. Overall, we are pleased with our gross margin performance, which reflects greater efficiency and scale in our Atlas business. We continue to expect that we'll see some modest reduction in overall company gross margin as Atlas continues to be a bigger portion of our revenue. Our operating loss was $10.2 million, or a -7 % operating margin for the second quarter, compared to a negative 15% margin in the year-ago period.
Our outperformance versus our operating loss guidance was in part driven by revenue outperformance, but also by the timing of certain expenses that we now expect to incur in the second half of the year. Net loss in the second quarter was $12.7 million, or $0.22 per share, based on 58.4 million weighted average shares outstanding. This compares to a loss of $0.26 per share on 55.6 million shares outstanding in the year-ago period. Turning to the balance sheet and cash flow, we ended the quarter with $975.4 million in cash equivalents, short-term investments, and restricted cash. Operating cash flow in the second quarter was negative $10 million. After taking into consideration approximately $5 million in capital expenditures and principal repayments of finance lease liabilities, free cash flow was negative $15 million in the quarter.
This compares to negative free cash flow of $13.8 million in the second quarter of fiscal 2020. I'd like to turn now to our outlook for the third quarter and full fiscal year 2021. For the third quarter, we expect revenue to be in the range of $137 million-$139 million. We expect non-GAAP loss from operations to be $27 million-$25 million. Non-GAAP net loss per share between the range of $0.48-$0.45 based on 59.1 million weighted average shares outstanding. For the full fiscal year 2021, we are increasing our revenue guidance to $549 million-$554 million. We are improving our profitability expectations and now expect non-GAAP loss from operations to be $71 million-$66 million and non-GAAP net loss per share to be in the range of $1.29-$1.21 based on 58.7 million weighted average shares outstanding.
Let me provide some incremental color on how we're seeing the ongoing COVID-19 pandemic impacting our outlook for the rest of the year. As we did on our Q4 and Q1 calls, let me start off by updating you on our underlying view on the business environment that's driving our guidance. In March, we expected COVID-19 to have a material impact only in the first half of the year. By June, we understood that the impact would extend into the second half of fiscal 2021. At this point, we clearly believe that the impact of the pandemic will extend at least through the end of this fiscal year. Starting with new business, our strong Q2 results notwithstanding, we do expect to see an impact on new business in the second half of the year. While customer engagement continues to be high, we expect macro uncertainty to be a factor.
Furthermore, our forecast assumes we're continuing experiencing slower than historic growth from our existing Atlas customers, consistent with our results in Q2. Finally, as you'll recall, our fourth quarter last year was a particularly strong quarter for our Enterprise Advanced product. Enterprise Advanced has an immediate impact on revenue due to the fact that under ASC 606, we recognize the term license component upfront. That dynamic makes Q4 in particular a tough compare for us. To summarize, MongoDB has delivered excellent second quarter results. We're executing well on our product and go-to-market strategies, which is generating strong growth at scale. We believe our clear leadership as the modern data platform of choice will enable us to continue delivering strong growth going forward. With that, we'd like to open it up to questions. Operator?
Thank you. We will now begin the question and answer session. To ask a question, you may press star, then one on your touchtone phone. If you're using a speakerphone, please pick up your handset before pressing the keys. To withdraw your question, please press star, then two. We ask that you please limit yourself to one question and one follow-up to allow everyone to ask a question. At this time, we'll pause momentarily to assemble our roster. Our first question will come from Keith Weiss with Morgan Stanley. Please go ahead.
Excellent. Thank you guys for taking the question, and congratulations on the very impressive quarter. Very impressive in terms of the momentum that you guys are keeping up is on both sort of the top line in terms of the nice growth that you saw there, as well as operationally in terms of those hires that you guys are able to get is both very impressive. One of the numbers in particular that I wanted to ask you about that really popped out to me was the number of new customer additions that we saw in the quarter. It looks like over 1,800 new customers. That's frankly the biggest number that I have on my spreadsheet here of you guys ever historically adding in what is admittedly a very difficult IT spending environment.
Dev, for you, is there any change in go-to-market or sort of a better motion like you were talking about in self-service? What's driving that strength and sort of that upward momentum that we're seeing in those new customer adds?
Yeah. Thanks, Keith. We're obviously very pleased with the customer additions for the quarter. I would tell you that we made some changes to sales incentives at the beginning of the year prior to the outbreak of COVID-19, where we made a conscious decision that we just wanted to accelerate the acquisition of new customers this year. We decided that because we saw how customers grew really quickly once they got onto our platform, and in particular on Atlas.
Okay.
Because we saw that growth happen, we said, "Gosh, why don't we just focus more on acquiring new customers, and they naturally will grow more?" Before we would, force may be a strong word, we would incentivize our salespeople to get upfront commitments. For many customers, they're not completely sure, one, of the platform, two, how quickly their workloads will grow, how much data they'll have, how many users they'll have. It ended up adding a little bit more friction to the sales process because they had to naturally try and convince a customer to make some form of commitment. We took that objection away, frankly, it's unleashed the productivity of our salespeople.
It's cut down the cycle time of some deals because avoiding that negotiation allows them to get on the platform more quickly and allows us both on the direct side and even on the self-serve side to just add a lot more new customers.
Got it. That makes a ton of sense. Then Mike, a question for you, and this is related to the guidance. The implied Q4 guide, if we take the full year and we subtract out the Q3, looks for a pretty sharp slowdown in overall revenue growth. I just want to make sure. It sounds like a lot of that is kind of like the tough comp that you have from strong EA a year ago period. Is there also sort of an aspect of mix shift in terms of the mix is expected to be so much more on the subscription side and sort of the new business coming to Atlas versus that EA, and that's also kind of impacting the year-on-year growth dynamic for total revenues as we get into Q4?
Yes. Thanks for the question, Keith. Overall, I think we have a very healthy outlook despite a challenging macroeconomic backdrop. As we've indicated, we certainly expect to continue to see or to see some impact on new business in the macroeconomic environment. We're also expecting to continue to see the Atlas consumption expansion at lower levels than we've had historically, consistent with Q2. Obviously, as you go later in the year, that also compounds. The last factor, which you did call out, is the tough compare. We called out last year in the Q4 that it was a particularly strong quarter for Enterprise Advanced. Actually even in that quarterly call, we talked about one large multi-year customer that represented about $3.5 million just from that one customer of Q4 of last year.
Just for those following the bouncing ball, not only do you lose that from the denominator or do you need to take that out of the denominator, but you also don't get the benefit of that in this year in the numerator. Overall, I think we are liking the dialogues that we're seeing. There's an incredible relevance, but there's macroeconomic uncertainty in the back half of the year.
Got it. Thank you very much, guys.
Thanks, Keith.
Our next question will come from Brent Bracelin with Piper Sandler. Please go ahead.
Thank you. One for Dev and a follow-up for Michael. Dev, I wanted to drill down on the record number of new customers as well and maybe take a slightly different spin here. On one hand, I certainly appreciate the self-service investments and focus there over the last year that you've made. That's clearly having an impact. It sounds like there's some incentives with the sales force that seems to be resonating and lowering some adoption barriers as well. Is there also an industry shift? My real question here is the product resonating in this current environment post-COVID in ways you didn't necessarily expect? Would you say most of the performance this quarter was all internal execution? Just parse how much of the industry is changing and the appetite for MongoDB is changing versus just internal execution, which again, was very strong.
Well, thanks for the question, Brent. I would say it's a bit of both. I would say that in general, what we're seeing is customers are really gravitating to modern technology stacks that allow them to be very nimble, move quickly, innovate fast. Obviously, MongoDB plays to that strength. Second, that they do want to focus on work that adds value to the business and basically outsource all the undifferentiated work. So managing a database is not exactly going to give them a competitive advantage. Third, given our platform offers them a lot of choice with offerings to allow them to run their workloads across the three different providers, that gives them also more confidence in terms of betting on Atlas. I think that's one clear dimension.
We have actually, I think as I said in the opening remarks, we put a lot of focus on our go-to-market efforts. In fact, I would say as much thought and effort we put into our product, we do the same on our go-to-market efforts. What I think you're seeing is just strong execution across our four channels. As I indicated in the last question, we have slightly modified, this is just a tweak, it's not a major modification, slightly modified some of the incentives to really make it easy for customers to engage with us and get on the platform quickly. That's really yielded the results that we've seen.
Certainly. Certainly something's working there and working well in the current environment. I guess, Michael, here, just for you as a follow-up. You talked about that COVID impact in the second half and gave a little more color on the tough compare on EA. As you think about the pipeline, would you say the pipeline of new business looking out in the second half is also slightly depressed, or are you assuming just a lower close assumption in the second half pipeline given all of the uncertainty out there? Just trying to parse out how much of your commentary on the business being impacted in the second half is tied to pipeline or just assumed close rates being lower given all the uncertainty out there.
Yeah, sure. No, totally understand. No, we feel very good about the engagement that we're seeing with customers. I think in the face of the macroeconomic environment, it makes sense to factor that in. We've got one quarter under our belt fully remotely, and I think we feel really well about how the team has done. The situation with COVID-19 and the overall kind of coordinated global slowdowns and openings and haltings is unprecedented. I'm incredibly proud of how people have adapted. This is not a playbook where you sort of said it, you've nailed it, and then you forget it. There's a lot operationally to focus on. Very proud of what the team does and how we've done. I think when you look out at the second half of the year, it's hard not to assume that the macro won't have an impact.
Very clear. Thank you.
The next question will come from Raimo Lenschow with Barclays. Please go ahead.
Hey, congrats from me as well. Two quick questions, one for Dev. If you start working with larger customers, there's obviously an application that usually sits there, and then the MongoDB database is the one that powers it then. What do you see in terms of people's interest? There's a lot of talk about digital transformation, but then with COVID-19, we have industries that obviously are not doing so well. What are you seeing in terms of that interest to do more stuff or realization we need to do more stuff, and then in effect, engaging with you on that? I would assume that's a multi-quarter story here that it should be starting to unfold. I'm just wondering if you see some of that already. I had a follow-up for Mike.
Yeah, there's no question that what COVID-19 has done is forced every customer, no matter how traditional or "legacy" they were, to be digital by default. Whether you're a traditional retailer, you have to quickly build out your e-commerce channel. If you're in a different type of business, you have to figure out a way to engage with your customers in a digital kind of format and approach. That obviously plays to our strength. One, customers can innovate and build very quickly on MongoDB. They can also scale it very well compared to all the other legacy platforms. Two, Atlas becomes very attractive because we take care of all the management of the database and the underlying infrastructure, so that makes it very attractive for allowing the customer to really focus on what's important to them.
Then, as I said, multi-cloud plays a big role, depending on what the customer's trying to do, what geo-coverage they need, what existing relations they have with certain cloud providers, et cetera. They have ultimate choice in terms of what underlying cloud provider to use with Atlas. All those three things play in our favor, and I think what COVID has done is really accelerated long-term trends. We were always very bullish about the future, but the long-term trends are clearly accelerating, and we're trying to capitalize on it.
Okay, perfect. Yeah, makes total sense. Then one for Mike. Mike, I hear you on the Q4 situation, but if I look at your Q3 guidance, it also kind of looks for kind of flat quarter on quarter, which is something we haven't seen. Is there anything in Q3 on the comps, et cetera, or EA that you wanted to point out?
No, I think in Q3, the key thing is we've got a very healthy outlook and strong customer engagement, but just a lot of macro uncertainty that we've incorporated into the forecast, which has been consistent with our general approach each quarter is to sort of mark the current state of the macroeconomic landscape as best we can to market and incorporate that into our thought process. From an overall engagement perspective, quite strong. We expect to see an impact on new business that obviously, particularly with the Enterprise Advanced product, has impacts in terms of revenue from term license revenue. We're assuming that Atlas is still expanding at below historic levels, consistent with Q2.
Okay, perfect. Thank you.
Our next question will come from Heather Bellini with Goldman Sachs. Please go ahead.
Thanks, guys, for taking the question. Just had a couple of quick ones. In regards to your comment about Atlas and the consumption, you're starting to see existing customer consumption start to come back. Is there a sense you could give us for how far off you think it is from where you were pre-COVID, and how long you think that takes to come back? I know that's a tough question given the macro environment, but I'll start with that one, and then I just had a follow-up.
Sure, yeah. Thanks, Heather. I would go back to our commentary on the Q1 call where we talked about it being sort of modest but broad-based and tracking the macro economy overall, where it was sort of cross-region, cross-sector, et cetera. That what we said is in the middle of Q2, we saw a rebound or an improvement in those lower levels that clearly, at least to us, when you look at all the data and we slice and dice it, indicates that it's tied to overall economic conditions. It's not quite yet back to historic levels. It would be very hard for me to predict when or how that would happen.
I think the correlations that we've clearly seen is it's broadly tied to macroeconomic behavior, which would make sense given both the broad-based nature of the behavior, but also the broad-based nature of the fact that it's a database application and it's a consumption revenue model, and so all those things should mirror each other. That's what we're seeing.
Okay. The follow-up question would be, just wondering from if your customers still might be asking for if you're seeing kind of a request for payment extensions or anything like that, or are you doing anything else creative on kind of top of funnel builds that might be helping down the line? I guess two other ones, sorry.
Yep, no, that's fine. We saw very limited and isolated instances of that in Q1. That has effectively worked its way through, taking care of itself. Those folks have come and cleared, and the frequency of requests or anything like that has really dried up. We saw a little bit right as the crisis was first sort of setting in, but it's not been a regular recurring sort of business consideration that we've had to deal with.
Okay, great. Thank you so much.
The next question will come from Brad Reback with Stifel. Please go ahead.
Great. Thanks very much. Dev, as you think about the evolution of the go-to-market and the acceleration in smaller Atlas customers upfront, coupled with a much broader product portfolio, is there any reason to think that your net dollar expansion rate shouldn't accelerate in coming years?
Yeah. I think just to make clear why we're doing that. We have seen Atlas customers grow very quickly. One of the reasons we want to get them on the platform is to really condition them so that they don't begin to think about even managing the next new app by themselves. Once they get conditioned on building every new app on Atlas, that becomes almost the standard way of people building and deploying applications. There's a long-term effect of that. It's hard for me to sit here in terms of what the implications will be on the net expansion rate. They've always been very healthy, and we don't see any potential change to that.
Great. Thanks very much.
Thank you.
The next question will come from Jason Ader with William Blair. Please go ahead.
Yeah, thank you. Dev, when you look at the database market, we're still seeing good traction for relational products, and I'm thinking of MariaDB in particular. Where do you think that traction is coming from, especially as the document model seems to be gaining significant share?
Yeah. What I would say is there's really two phenomenons. There's what I call a phenomenon of lift and shift, where customers just want to get off the legacy vendor, in most cases it's someone like Oracle, where they don't want to pay the Oracle tax anymore, and go to an open source relational database. That's what many people call the lift and shift phenomenon. Many customers naively believe that that will allow them to just keep running their business. What they miss, though, is that they miss the opportunity to refactor that application and really modernize it to position themselves to really take advantage of all the new capabilities of a distributed platform and take advantage of other things like microservice and so forth.
We have many customers who decide, you know what, rather than just lifting and shifting, we want to re-platform and basically choose MongoDB as the destination solution. It really depends on how quickly the customers want to move, their perceptions of how difficult it is to move off their, say, Oracle, to a distributed database like MongoDB, how enabled they are, and also how much time pressure they're in. That's why we're expanding our reach, working with systems integrators, working with our partners, and obviously our own sales force to get out there and really explain the benefits of MongoDB and the fact that whatever perceptions they have, we can address to make the choice of making MongoDB that much more easy.
Do you think the current environment will change those dynamics at all in terms of those phenomenon? In other words, maybe move people faster to the modernization of re-platforming phenomenon?
Well, I think when we first came out, a lot of people viewed us as an interesting but niche solution, maybe just used for a certain set of use cases. I think we've clearly shown that MongoDB can serve the most demanding customers anywhere in the world for a wide variety of use cases, and that's only increasing. I think people are recognizing, and as I mentioned, all the major players today have had to acknowledge that JSON is the format and the structure that developers like to use, that the document model is truly the best way to work with data, and the added benefits of our distributed capabilities. It's so easy to scale out using a distributed platform like MongoDB than, say, staying on a relational platform and having to do a lot of gymnastics to get the application to scale.
I think it does take time, but we've clearly seen customers recognize and use us for very sophisticated applications.
Thank you.
Thank you, Jason.
Again, if you have a question, please press star then one. Our next question will come from Tyler Radke with Citi. Please go ahead.
Hey, thanks, and good evening, Mike and Dave. Michael, I just wanted to start with you in trying to understand the accelerating customer adds here. It sounds like you did make some tweaks to the sales incentives to prioritize adding new customers. I wanted to just better understand the financial impacts here. Are you landing new customers at a lower ASP and then hoping that those expands are higher down the road? Maybe just help me understand that, and then how we should think about just the accelerating new customer adds translating into higher revenue growth, or just how to think through the financial impacts of that.
Sure, yeah. There are a couple different factors and flavors of it. You can slice it by channel or you can slice it by product. I'll try and do both relatively quickly. If you look at it on a channel basis, it's really just improvements in the returns from the investments that we've been making in the self-serve. Self-serve customers, as you know, come on at $5,000 or $6,000 in annualized spend. The unit economics are very attractive of that given the cohort behavior. As Dev mentioned in the prepared remarks, we're continuing to get better in terms of the muscle and introducing more of the DNA there. We've been pleased with the results that we've had, and so that shows up in self-serve customer additions.
On the direct sales side, we've also made some sort of refinements to making it easier for customers to use the platform. As Dev was mentioning in response to one of the earlier questions, rather than a salesperson being focused on maximizing the magnitude of the initial commitment, we know based on the data and the cohort behavior for Atlas customers, that getting them on the platform is the most critical thing, because then usage will expand from there. So we've decreased some of the customer-related friction in terms of getting folks on the platform, which makes it easier for people to adopt. Then reps getting paid for the consumption over time, rather than try to get paid on some maximum initial commitment. We're also seeing really good success there.
To the revenue question, customers will start off or may start off consuming less. I think from a revenue perspective, it doesn't really matter, because even if someone made a big commitment, their initial consumption still is small. From a revenue consumption perspective, the dynamics will generally be the same. It will be different in terms of how much do we bill up front or some of those things that aren't areas that we focus on, but they're things that sometimes investors focus on. That's one of the other reasons why we call it the dynamic, just so people can understand that.
Okay.
Most of what we're talking about, just back to the product view of things, most of what we're talking about really is sort of Atlas related and sort of Atlas centric. There's certainly in any given quarter, mix and variability between what the ultimate list of deals looks like in a given quarter. The piece that we're talking about here are really Atlas centric.
Right. The acceleration kind of is mainly coming from Atlas, which I think you can see.
Tyler, what I would also add is that it's not lost on us, and that in difficult times, most companies gravitate to selling to the existing install base. As we've talked about many times in the past, we do have a land expand strategy, so our sales force could easily do that. We thought just myopically focusing on existing install base, would be a mistake because we do see a big market opportunity. The added benefit of engaging with them earlier is that you just condition them to start using Atlas as the default platform by which they build their applications. That ends up building a long-term durable relationship with those customers.
Yeah, that makes sense. Just a follow-up for you, Dev. Would just love to get your view on Oracle's autonomous JSON service. I know you briefly touched on it in the prepared remarks, but curious if you have a view on why you think Oracle introduced the service. Do you think it's in response to greater competitive pressure that they're seeing from you guys? Would just love to hear if you have a view on the pricing claims from Oracle that it's 30% cheaper.
Yeah, sure. Couple of points. One, they provided support for JSON in 2014. What they recently announced was support for JSON and on their autonomous database in the cloud. One, it only runs in their cloud. Two, much like what Microsoft did with Cosmos and AWS did with DocumentDB, it's really built on top of a relational database. It's basically a shim that basically tries to allow developers to more easily use JSON, but the back end is basically the same old engine. There's some severe feature limitations. One simple example is that because it's basically a JSON blob, you really can't even parse numeric data. It's all basically text. There's some severe limitations in terms of how you can use those features. Obviously, because it's built on a relational back end, there's also a huge set of implications around performance.
To the latter part of your question, this is really in response to the fact that developers are recognizing and are acknowledging that it's so much easier to use JSON or by definition, documents to work with data. It's so much easier to code. It's much more natural. It's much more intuitive. It's much more flexible. Oracle basically acknowledged that. We believe that as the world moves more and more towards using documents, we are well-positioned. Obviously, we've been competing with clones for quite some time, and I think we've done quite well.
Thank you.
Again, if you have a question, please press star then one. Our next question will come from Jack Andrews with Needham & Company. Please go ahead.
Good afternoon. Thanks for taking my questions. I wanted to ask a couple of product questions. First, I was curious if you could talk about what you're seeing in terms of the uptake of search so far within your customer base and any context around changes in consumption patterns.
Yeah. Just so we're clear, we made these products generally available in June of this past year. It's very early days. We're quite excited about the uptake so far. The reason, just to remind everyone why we introduced search, is we actually had customers telling us, "We much rather stay on one platform than to have to go learn, build, and maintain a separate platform." Our Atlas Search product is obviously fully managed as well, and you can take advantage of all the features like MQL that you could before with the regular database server. It's still early days, but we're very pleased with the results to date. As we get more focus and attention from the sales force on this, we feel quite good about the trends in the business.
That's great. Just as a follow-up, how should we be thinking about the monetization opportunity around Realm? Just trying to get a sense for what proportion of your customers do you think need the full capabilities, between Realm and with the cross-sell with Atlas?
Yes. The reason we acquired the Realm Mobile Database and now have built a tight coupling between the mobile front end and the Atlas back end is to really enable developers to take advantage of a much more powerful format. Realm's architecture is very similar to the MongoDB architecture. Mobile developers find it very useful to use the Realm Mobile Database. Realm itself is not meant to be a revenue generator unto itself. It's really to enable mobile developers to think about using Realm and Atlas first because it allow them to move that much more quickly. What you'll see, in terms of revenue impact, will be more on the Atlas consumption at the back end versus us trying to monetize the front end. Typically, the mobile database, the client side, is given away for free.
Our approach is to really enable developers to build very sophisticated applications. One of the key features, Realm Sync, which is still in beta, really simplifies the work a developer has to do to be able to synchronize data between the client side and the server side. As you can imagine with people using mobile devices or other devices that are not tethered to a direct network, there's always a chance that you could be offline. Being able to quickly synchronize data between the two, between the front and the back end, become very important. Developers have to incur an enormous amount of complexity and cost to be able to do that manually. Offering this capability will make a developer's life that much more easy and will make our platform that much more compelling.
Got it. Thanks for the color, and congratulations on the results.
Thanks, Jack.
Our next question will come from Pat Walravens with JMP Securities. Please go ahead.
Great. Thank you. This is Joey on for Pat. Dev, just one high level one for you. I wanted to touch on the employee engagement survey you mentioned. How do you make MongoDB a place people want to work at and then keep them engaged, especially in the current work from home environment? Thank you.
Well, thank you for that question. For a business like MongoDB, our success is directly correlated to our ability to attract and keep great employees, both on the product and engineering side, as well as on the go-to-market side, and obviously, as well as the people who support the entire business. We were very intentional when COVID-19 happened to really be very transparent with employees about what we were doing. Now, candidly, we were in a very healthy financial position. While we did worry that there might be some short-term impacts to business, we weren't changing our posture or potentially laying off anyone. I think there's two questions that have impacted anyone in COVID. One, do you still have a job? Two, can you do your job safely? Those two questions at MongoDB have been an unequivocal yes.
More importantly, we really communicated to all our employees about making sure they were clear on their priorities, making sure that our leaders were clear with their teams about what was important, making sure that they understood where we're going in terms of what we're trying to do, making sure that we made it very easy for them to work from a work from home environment without much disruption. We've also been sensitive to the fact that working from home for many people is very challenging, especially if you have young kids at home or you don't have ample private workspace at home. So we've been very flexible with employees to make sure that they can take sufficient time off, work flexible hours, et cetera. I think we also really live our values.
That's another way of saying we really try and create a culture where people feel like that this is a great place to work, that they're respected no matter who they are, that they can not just have a job, but a career, and that we want to set them up for success. We try to do that. We're very grateful for what our employees have done since COVID. Our results are obviously a reflection of that. We've been hiring aggressively. We're very focused on onboarding those new employees to make sure that they get quickly inculcated into the culture.
Awesome. Thank you. Congrats.
This concludes our question and answer session. I would like to turn the conference back over to Dev Ittycheria for any closing remarks. Please go ahead.
Thank you for joining us today. I just want to wrap by saying that I'm really proud of the strong execution this quarter in a difficult macro environment. We've also expanded our product portfolio to really increase our lead as the preeminent data platform on the developer. As we discussed, we've had a record direct and self-serve customer additions demonstrating that our strategy of adding new customers is working. Lastly, we're pursuing a big market opportunity, and COVID is only accelerating long-term trends. Thank you for your time, and we'll talk to you soon. Take care. Bye-bye.
The conference has now concluded. Thank Thank you for attending today's presentation. You may now disconnect.