Full Transcript
Jeremy Levy: This is Deciding my Data. I’m Jeremy Levy with Andrew Weinreich. Our guest today is Meghan Gill, who is an award-winning marketing leader and startup adviser. She was employee number eight and the first non-technical hire at MongoDB, where she built a developer community from the ground up and launched many of the company’s first programs. She now runs sales ops at MongoDB. Meghan, thank you for being here.
Meghan Gill: Thanks for having me.
Jeremy Levy: Before we get started, it would be great if you can give us a quick overview of MongoDB as a company and also what actually is MongoDB, the database.
“MongoDB has been downloaded over 30 million times.”
Meghan Gill: Sure, sure. So MongoDB is a new kind of database technology and it’s primarily used by software developers. So those are the people that are building all of the amazing modern applications that we use every day. Since it’s Deciding by Data, I’ll give you some stats. So about 18 million developers worldwide according to IDC. And MongoDB has been downloaded over 30 million times. So you can sort of see the magnitude of adoption of MongoDB. Some other interesting things about our business model, so, we kind of have a, something like a premium business model and that MongoDB is open source which means it’s free to download our community edition and then we sell an enterprise version that has a set of features that are- we monetize a set of features that are more relevant to the enterprise, things around security and management. And we also have a hosted database product called MongoDB Atlas.
Jeremy Levy: Could you give us a background also on your role, I know you’ve been, you’re obviously employee number eight, you’ve been there for a very long time. Can you tell us about your role and how this evolved over time?
Meghan Gill: Sure. So when I joined roles were pretty loose, as you can imagine on a eight-person team. So everybody was sort of doing a little bit of everything, but I was always pretty focused on marketing, particularly marketing to our developer community. So we have people all around the world who are using and trying and experimenting with MongoDB. How do we reach them and how do we get to be successful? Over time we built a full marketing organization, it was pretty sophisticated, very data-driven. And I ended up running developer relations, demand generation and events, which was a lot of fun. And then, most recently I took on a role, a totally different role, which is also very data-driven, running sales operations, which is all the tools, systems reporting and analytics around are our enterprise sales organization.
Jeremy Levy: I want to get into the details of how you use data in a second, but when I think about databases, I think about columns and rows and relational and SQL. What makes MongoDB so special that- and you mentioned NoSQL a second ago, can you expand for the audience a little bit what actually that means?
“90% of data has been created the last two years, which is pretty mind-blowing. And, 80% of the enterprise data is unstructured, meaning it doesn’t fit into the tables that are in a relational database.”
Meghan Gill: Sure. So traditionally, databases have been very much like you described; rows, columns. And MongoDB is really a different approach. It’s really built to address the modern requirements of today’s modern applications. So putting things in context, 90% of data has been created the last two years, which is pretty mind-blowing. And, 80% of the enterprise data is unstructured, meaning it doesn’t fit into the tables that are in a relational database. So MongoDB is a different approach, it’s more flexible, more scalable and really meets the requirements of having a lot more data that’s also unstructured.
Jeremy Levy: So was that the original vision for MongoDB back when the company first started or was that sort of an evolution of startups and learning and seeing what works and what not?
Meghan Gill: So our founders came from DoubleClick, so they were dealing a little with tremendous scale. And I think, their thinking was they wanted to create the database they wish they had at DoubleClick and they had built plenty of custom data stores that were doing similar things and MongoDB is sort of an outgrowth of that experience.
Jeremy Levy: And so, in the original sort of marketing plan, what was the plan in terms of distributing or getting organic growth around MongoDB? I mean, I always think of this as sort of a great story around developer evangelism. Tell us a little bit how that started and also how are you even tracking that and thinking about metrics and data perspective.
Meghan Gill: Yeah, it’s funny when you ask about this period in MongoDB. It was in the very early days, very difficult to be data-driven. And I think that’s true in a lot of startups because you don’t have a whole lot of data, you don’t have customers. And in our case, we were really focused on adoption in the early days, getting developers to get their hands on the open source version. And we didn’t even put a registration wall behind the open source edition in MongoDB. So we had maybe an IP address that gives us some signal about where the person is, maybe even a company but we don’t know specifically who they are. So trying to find signal early on was a bit of a challenge and a lot of the programs that we run were really focused on meeting developers where they are. So developers are on social media, they’re on forums like Hacker News or on Reddit, they’re at Meetups and events like OSCON or PyCon or the other sort of developer-focused events. So we made a big effort to be in those places with either content or speakers or engineers to talk to them about how MongoDB would be a better way to store and manage their data.
Jeremy Levy: Was the developer-first approach something that you had sort of an intuition about or was that something that had evolved also? Was that something that perhaps the numbers that you were tracking from an insulation perspective were showing you.
“Developers are the new kingmakers, they’re the ones that are making decisions about the technologies that are used.”
Meghan Gill: Well, I think one of the challenges with a database is that you have some very big entrenched players. So I think we all know which companies that I’m talking about or thinking about. And, those companies have a pretty big presence with the CIO, the C-suite. So, MongoDB kind of took the approach of like, okay, rather than going top up, top down where we have a bunch of competitors, let’s go bottom up. Developers are the new kingmakers, they’re the ones that are making decisions about the technologies that are used. And it’s true, lots of apps get built by an individual developer or development team within an organization and then it bubbles up to management and next thing you know there are multiple apps within a company and that’s where we are able to sell an organization our enterprise product.
Jeremy Levy: But was that something that you had said this is the strategy we’re going to take? We’re going to go for engineers and because more and more decisions are coming sort of from the bottom up? Or was it something that you would sort of observe through maybe some of the metrics that you were tracking that said, “hey, wait, we’re seeing a lot of activity with engineers and we should double down that tactic.” What I’m trying to understand is, is that something that you guys started off thinking about or did you evolve into that as a tactic, as a strategy, rather?
“So our first two courses were MongoDB for developers, MongoDB for database administrators. And we had no idea that thousands of people would sign up within the first week. And now we’ve had over 500,000 people take an online course on MongoDB.”
Meghan Gill: I would say that we started off that way and then data reinforced the decision. So, I mean, as an example, I mean, the numbers of downloads that we were getting, those were all from developers. We launched online education, we offer free courses for people who want to learn MongoDB. So our first two courses were MongoDB for developers, MongoDB for database administrators. And we had no idea that thousands of people would sign up within the first week. And now we’ve had over 500,000 people take an online course on MongoDB.
Andrew Weinreich: What was the promotion you did for those first online courses? How did you get the word out?
Meghan Gill: We did a lot of typical marketing campaigns. We did email marketing, social media. We did press around it. We partnered with EdX to deliver the courses. It was interesting, we did an online event where we were talking about the new features in the upcoming release. And, at that event, we announced that we were offering these free courses on MongoDB. And the response was very viral because people were really craving the education about this new technology, so being able to deliver it in a scalable way has been pretty incredible.
Andrew Weinreich: We’re always curious about missteps. It doesn’t sound like there are missteps here. I mean, it doesn’t sound like there was an initial thesis that the data informed you, you needed to pivot. It sounds like you were dead on with who the audience was and then you were dead on with how you would promote that initial educational program. Do I understand that correctly or is there a misstep that you can share with us or a misdirection or some initial thesis that was disproven?
Meghan Gill: Well, I do think our initial thesis is to go after developers but there has been an evolution because when we sell, when our sales team goes and sells, they don’t just sell to developers. They have to get the operations team on board, the architects, often there’s a line of business. So I think initially we were very focused on the developer but then as we built an enterprise sales organization, they were like, “Hey, we need a different collateral, different messaging, a different way to approach these other personas.” So, I do think that was something that evolved over time. But typically the way that the sales reps would get into an account would be to see that there’s what we call in MongoDB parlance we call it smoke. Is there smoke in the account? Is there’s stuff going on? There are teams that are downloading MongoDB, are they building applications? Then, they can go take that information and they have a message that they can tailor to somebody higher up in the organization or talk to them about MongoDB enterprise.
Jeremy Levy: Has the business model also evolved? I mean, early on, I think you mentioned a second ago, it took a similar model to the Red Hat model where we’re giving away this offer for free but the consulting and support is where the relationship comes in. Has that evolved also?
Meghan Gill: Yes. It’s evolved in two ways. The first way is that we built an enterprise edition, so we actually have a different set of software. That includes security, tooling, management and a whole bunch of other things that make running MongoDB production much simpler, much easier. And particularly for the operations team that have to manage very large MongoDB clusters. So it’s beyond- that also includes support, but it’s beyond sort of the support and consulting that some other open source companies provide.
And then the second thing I would say is that now we offer a cloud product. So if you want to run MongoDB in the cloud hosted by MongoDB, we have a product called MongoDB Atlas. An interesting thing about that is that we also have a self-service channel. People can just go swipe their credit card and spin up on MongoDB clustering using Atlas, which is pretty amazing. And our sales team sells that as well. So we have sort of monetized, we had this big open source community, people learn MongoDB for free using the community edition and we can either monetize them through the enterprise edition or through our cloud version.
Jeremy Levy: So, I want to I want to talk a little about where we are today, but before we sort of leave this section we’re talking about the past, one last question. Aside from downloads, what were the other things you were tracking that you were using, whether it was smoke for signals or other things that you were thinking about in terms of the metrics perspective that you were tracking, to help understand where to focus? At least in those early days.
Meghan Gill: Yeah, in the early days when we had less signal, I mean it was things like, there were some qualitative things, like what kinds of questions people are asking about MongoDB, we were looking at things in support forums like Stack Overflow, we were looking at GitHub, so which language communities are engaged with MongoDB, downloads, people putting MongoDB and other technologies on their LinkedIn profiles. So there was, there were definitely signals and of course, we did start building the demand engine that captured people’s information, started to educate and nurture them based on what they sort of interacted with. But in terms of making decisions about how we should approach marketing and approach product, approach sales, a lot of that came from those downloads and the other signals like that.
Jeremy Levy: So as the product portfolio has evolved today, you mentioned the platform also, are there other sources of data that you’re now leveraging in terms of thinking about how you’re marketing it to your customers? So you mentioned you’re now capturing data, as this evolved you’re now capturing data for sales perspective. You’re selling multiple products now. Are there different metrics that you’re now tracking in terms of how you are thinking of making decisions from sales and marketing perspective?
Meghan Gill: Sure.
Jeremy Levy: Other than revenue, of course.
Meghan Gill: Yeah. I mean, we have a pretty sophisticated demand generation engine, so we have, we obviously have a serum, we have marketing automation in place. So we’re running multiple campaigns, we’re tracking the responses to those campaigns, the budget for those campaigns and then running quarterly reports to understand, should we be investing in a particular channel, whether it’s events or collateral or webinars or a particular theme or topic. Is it microservices? Is it mainframe offloading? Is it blockchain? What is resonating with our audience? So now that we have, I think we used over 20 different marketing technologies, that helps us sort of get a better picture of what is compelling to our audience and also feed them the right content. So we use Eloqua. We have 60 plus tracks running in Eloqua which are all triggered based on people’s persona, their industry, their engagement with MongoDB. We use a tool called Demandbase which lets us personalize the experience on the website based on the IP address and whether they’re an enterprise company or a startup or what the location is. And those are just a few examples of how we’ve been able to make a more personalized experience for our prospects.
Jeremy Levy: How much of that would you say has helped inform- or rather how much of sort of collecting additional information, the amount of the different tools, how much of that has allowed you to get much more specific, maybe from a predictive perspective on what leads to go after, what channels to go after? Does that make sense, the question?
Meghan Gill: Yeah, I mean one of the challenges we have is that because we have such a big, it’s a champagne problem but we have a very big developer community and so a lot of times our sales organization will get like will which accounts do I focus on? The other thing is that MongoDB is a broad platform. Any company and any software development team could use MongoDB. So other products might be more focused on a smaller persona or on a specific vertical but MongoDB is very, very broad. So, in terms of helping to make decisions about which accounts to go after, I mean, we are now using predictive lead scoring that helps us better prioritize the low leads and the MQLs and appropriately score the MQLs. And then we’ve also put in place some tooling that we built sort of internally to help our sales team figure out which accounts they should target and help the sales managers manage territories so that they’re figuring out how to distribute the accounts we call the smoke, across the reps in sort of an equitable way and assign the right accounts to the right the reps based on the reps experience and based on the activity in the account.
Jeremy Levy: What learnings would you say you have for other startups that can take away from your experience when they got to really have the same amount of data that you guys have now?
“…we historically collected a lot of data and put a lot of effort into gathering the data in the systems and I think that’s benefiting us now.”
Meghan Gill: Yeah. I think it’s, well, one of the challenges, I think, at any startup is you want to move quickly, but at the same time, you want to set yourself up for having tremendous success. So sometimes you make decisions about the systems and the tools because you’ve got to get stuff done and then you may be stuck with these decisions. So, I think there’s always a delicate balance. I mean, we historically collected a lot of data and put a lot of effort into gathering the data in the systems and I think that’s benefiting us now. I think as we get bigger — MongoDB’s over 800 employees now, so pretty sizable company — I think one of the challenges we have is just stitching all that data together across the different systems and tools that we have.
Jeremy Levy: Where do you think that you’ve iterated your way to success as opposed to sort of getting it right out of the gate? I mean, where have you sort of said, where would you sort of focus in terms of where you’ve improved over time?
Andrew Weinreich: Both on the marketing side and on the products side, maybe even to a lesser extent on the product side. On the marketing side, would be super interested to understand where you see that iteration.
“…the other amazing thing about an open source product like MongoDB is … the users that are used to giving feedback, submitting tickets, we have tickets that community users have been voting on, and that has helped inform our roadmap pretty significantly.”
Meghan Gill: I think certainly in how we run demand generation, we’ve got a lot of iteration, so as I sort of mentioned, we, we were very developer-focused in the beginning and now we run programs to target all the different types of personas that are relevant to MongoDB. And that was definitely an evolution because I think we got very used to talking to a developer and the needs an operator or an architect might be quite different.
And then, in terms of product, I think the other amazing thing about an open source product like MongoDB is that we get a lot of feedback from, that the users that are used to giving feedback, submitting tickets, we have tickets that community users have been voting on, and that has helped inform our roadmap pretty significantly. So, I mean, I think in both marketing and product, there have been areas where we’ve iterated.
Jeremy Levy: Switching back to the product for a second, Meghan, I’d be remiss if I didn’t ask about either AI or ML when it comes to sort of big data ecosystem and Mongo playing a role in that. Where do you sort of think about those spaces and where do you see MongoDB going within the AI and ML sort of category?
Meghan Gill: Well, when I’ve talked to our founder and our CTO about AI, his insight is that a lot of these algorithms have existed for a long time, but we finally have lots and lots of data that we were collecting that were able to run those algorithms through to make them more useful. So I think MongoDB being a flexible database can power a lot of these applications. For example, there’s a company called x.ai, that’s a personal assistant, AI-powered personal assistant that’s built on MongoDB. Many of the predictive analytics and social media analytics companies have MongoDB, are rebranding MongoDB to sort of manage all the unstructured data. So I think that’s sort of where it fits into that space.
Jeremy Levy: Is that more of an area where you see an opportunity for Mongo to play a role? The big data aspect as I think about it is, I think what you mentioned was that Mongo was a repository for this information. Is that more and more of a use case that you’re seeing Mongo deployed to? And, is that an area you are pursuing from a sales and marketing perspective?
Meghan Gill: I would say yes and yes. We do have companies that are using, some customers are using MongoDB to build AI-based applications, to build big data predictive applications. So is it a place that we’re pursuing? Yeah, we have built out some content and we’re sort of testing out if that’s a draw for people. I think companies are still sort of figuring out how AI sort of fits into their broader strategy.
Jeremy Levy: Are there other trends in your ecosystem that you see for either future uses of Mongo or future products that you guys are thinking of building around that you can speak to?
Meghan Gill: Yeah, sure. I mean, there’s like a few technology trends, for example, one might be microservices. So organizations are sort of moving from having a single monolithic application to having many smaller applications, and MongoDB could be a good fit for that. Another trend might be, there’s a new technology called blockchain that’s pretty interesting. I’m trying to think of the other things that are really resonating with our audience. Obviously, cloud. So we have a cloud-based product. I think many organizations are figuring out what their cloud strategy looks like and that might not be straightforward. But now that we have MongoDB as a service in the cloud we’re a big part of that conversation with a lot of our customers.
Jeremy Levy: What do you see in terms of the future from a marketing perspective?
“I think that over time, I think our cloud product is going to become more and more important.”
Meghan Gill: Well, I think that over time, I think our cloud product is going to become more and more important. And, the other interesting thing about the cloud product is previously we had this open source version, people could use it for free but they had to sort of set it up, manage it on their own, and we didn’t have a way to monetize it. Now by offering it in the cloud, we are able to help people get up and running much more quickly. Developers don’t need to deal with sort of management aspect of MongoDB, and we’re monetizing them. So I think that that’s sort of the future for us in figuring out the marketing strategy around that sort of self-serve channel is really important as well as enabling our sales organization to actually go out and talk to customers that are thinking about their cloud strategy and help sell them Atlas basically.
Andrew Weinreich: What is the marketing strategy around a self-serve channel in the future. Is it substantially different than it is now? How does it evolve?
Meghan Gill: Well, I think we’re targeting- previously we didn’t have a way to sort of monetize developers directly or we didn’t have a way to help them manage their applications easily. So when we talk to developers we try to get them to use Atlas first. There’s a free tier that we try to get up and running very quickly. So, that’s sort of a shift from how we talked to our primary audience. Rather than download MongoDB community, get started, we’re like, “Hey, this is an even easier way for you and for us.” It sets us up as a company for them to be using our cloud product. So from a marketing standpoint, a lot of the programs that we run, if we’re showing someone how to use MongoDB we try to show them how to use it in Atlas, at an event or a webinar, whatever it might be. Run lots of digital programs and doing lots experiments on various digital channels to get people to sign up sort of directly for Atlas. You know, developer evangelism team and they are also sort of speaking and talking about MongoDB and they’re typically doing it through, like, here’s how you would do it in Atlas. So I think that that’s the shift.
Jeremy Levy: Staying on Atlas for a second, does that mean you are going to be competing head to head with an AWS or Google Cloud platform with regards to offering a quote, ‘database as a service,’ in the future or expanding those toolsets as well?
Meghan Gill: So Atlas is a database as a service and we actually partner with AWS, with Google, and with Microsoft Azure. So we offer it on their platforms. So it’s more of a partnership with those cloud providers. So one of the nice things about Atlas is that you have a choice. If you want to run your database in the cloud, you can run it on any of those public cloud providers.
Jeremy Levy: Oh, I see, as part of their sort of marketplace offering, where with one click you can spin up a third-party vendor’s platform, so to speak, running on their raw EC2 instances or their droplets and so on.
Meghan Gill: Yeah. So the idea is that with Atlas you can use whatever cloud provider that you want and you can get a fully-managed database as a service.
Jeremy Levy: Cool. Well, Meghan, thank you very much. This is Deciding my Data. I’m Jeremy Levy with Andrew Weinreich. Thanks very much, Meghan. Really appreciate it.
Meghan Gill: Thank you.
Andrew Weinreich: Thanks, Meghan.