Episode 42: EnterpriseDB, Collaborating with the community to make Postgres enterprise ready, with Ed Boyajian, CEO
Ed Boyajian, CEO joined EnterpriseDB and helped it pivot from a small organization, to one of the leading Postgres database companies. The company has figured out how to run a profitable business, while embracing and respecting the community and open development process that has formed around Postres for more then two decades.
Michael Schwartz: Hello, and welcome to the first episode of Open Source Underdogs in 2020. I’m your host Mike Schwartz, and this is episode 42 with Ed Boyajian, CEO of EnterpriseDB.
This episode was recorded last year, it took me a while to get it out due to some technical challenges. Let’s just say the internet Gods conspired against me the day I recorded this episode, and I had to finish recording on Zoom, which is fantastic for meetings but not ideal for podcast. So, if you hear a slight audio quality difference towards the end, that’s the reason.
After the episode, I have an announcement about what you can expect this year from The Underdogs podcast. We have a new challenge, but I think you’ll be excited to hear about it.
So, make sure you tune in after the main event.
Ed is definitely one of the superstars of open source business. I got a lot of great insights from this interview. And I feel like I only scratched the surface, we might have to have him back for a follow-up.
But, without further ado, here we go. Ed, thank you so much for joining the podcast today.
Ed Boyajian: Hi, Mike, I’m glad to be here.
Michael Schwartz: What was the EnterpriseDB product when you joined a CEO, and what were your first priority to transform the business?
Ed Boyajian: I came to EDP in 2018, and the company had been founded a few years before that, in 2005. The original thesis for the company was centered on helping customers solve problems they were having with Oracle. And even then, there was well-known pain around locking that was associated with Oracle and EDB, that is origination, and had developed some technology to make compatibility with Oracle, kind of the flagship technology, such that it was easy to migrate applications, written to run on Oracle, to run those on a Postgres database.
So, when I joined the company, that was kind of the center point of the business. And a company at that time was more prominently known as the Oracle compatible database company. My focus when I joined was to really shift that and center it much more squarely on it, being a Postgres database company.
Michael Schwartz: What were some of the biggest challenges you faced when you joined a CEO to pivot the business?
Ed Boyajian: I think, like many companies that have to face change, the core engineering focus of the business, being on Oracle compatibility more so than Postgres, meant that there had to be a shift in mindset, and to get people to really reorient around prioritizing Postgres and our competencies in the database platform itself, as the priority for the company.
And having come from Red Hat, I’ve been there for almost seven years and had some experience obviously around open-source projects, and working with open-source communities, I think that was an important change for the company to really think about what it meant to be participants and contributors to a project like Postgres.
Relationship with Postgres Community
Michael Schwartz: Did the Postgres community welcome EnterpriseDB? How would you describe your relationship with the community?
Ed Boyajian: I think that the Postgres community was and is very welcoming of contributions from outside, and to that end, and at that time, very, very welcoming to EDB in our contributions.
I think the company had done work with Postgres previously, but in the intervening time, we’ve invested in a number of people, who’ve come to the company, that are key contributors to the community. So that’s been a real strength.
Postgres, like Linux, as you may know, is an independent open-source community. It’s not one that’s controlled by a company. I think that’s inherently the strength of the community. And I think, given the longevity of the Postgres community and the nature of governance, the Postgres is governed by a core team of folks that manage the whole project – there’s no single individual who’s responsible. I think that is also structurally aligned well to encourage contributions from companies like EDB.
Open Source V. Open Core?
Michael Schwartz: Cloudera and Chef recently went from open core to 100% open source. Of course, you’re familiar with Red Hat that has a similar model – do you think that open-source vendors are moving away from open core?
Ed Boyajian: I see generally that being true, although many companies, including EDB, do work on the periphery, around the core, open-source project. We do that in the form of tools and extensions that aren’t specifically in the database server itself, and we do those projects outside of the core community.
I think a lot of companies, including Red Hat, do work in similar fashion, but generally speaking, I think there’s a great advantage to being prominent, an active contributor to project. And I think most of the commercial companies associated with open-source projects have recognized that value.
EDB Value Prop
Michael Schwartz: Speaking of value, what are some of the most important value propositions for EnterpriseDB?
Ed Boyajian: I think there really a couple that we focus on, and I think that means most to our customers. The first is around customer obsession, but it’s customer obsession set entirely in the context of Postgres.
And to kind of put another lens on it, I think it would be fair to say Postgres One. That enterprises and governments all over the world have made Postgres a standard database that’s now part of their strategy.
EDB was central in creating that when in the market, and because of that I think, we’re uniquely positioned, with our exclusive focus on Postgres, to provide a level of care for customers that no one else in the market can do at this stage. I think that’s one.
The second is Postgres technical differentiation. And for EDB that manifests in a couple of ways. Primarily in our contributions to the community, where we’ve led advancements in the core Postgres database platform for many of our customers, parallel query, projects like zheap, and other projects, where there’s a meaningful amount of heavy lifting, we’ve been the primary contributors to much of that work.
The second area where we create technical distinction is what I mentioned earlier in the enhancements, that we make kind of around the database server itself. And some of them touch the database server like Oracle compatibility, which many of our customers still value from EDB, but beyond that, doing work in the areas that enable Postgres to be deployed at scale have been really important.
You can think about that in the context of replication, or failover, enterprise-class management, and monitoring – all the things that any enterprise would need to run a database at scale.
Michael Schwartz: How many EnterpriseDB products are there today? And which are most important from a revenue perspective?
Ed Boyajian: So, predominately, we think of it as essentially being one primary product we called the EDB Postgres platform. Within that, there are two database server options: one is the pure community edition, the second is that version of the database server that’s enhanced with the work we’ve done in Oracle compatibility, and in a few other areas.
We bundle with the tools that I described earlier, such that our customers don’t have to pay for a separate skewer line item for the pieces and parts they need to make the database run. And that’s notably different than the traditional enterprise vendors in the market.
Michael Schwartz: Ed, pricing is really hard for startups – do you have any advice on how to find the right gates, or how to set the right price, or how to evolve pricing as the business environment changes?
Ed Boyajian: That’s a great question. It is a part of the strategy that’s evolved over time. In the early stages, as we started to develop some of those capabilities around the database server itself, we attempted to monetize those as parts.
And we found that that was particularly complicated largely because customers were turning, the Posgres were turning away from complex, contractual arrangements and complex, buying systems they have with Oracle and other commercial vendors.
So, I think our first attempt at that we got wrong, by putting that into a bundle, it also created the dilemma of figuring out how to price that sufficiently, such that we didn’t over-price. Because we had a lot of extra development, and IP, and value-add on the one hand, but on the other hand, kept things simple.
We, over the years, have been really tuned in kind of creating price points that are appropriate for customers. And I would say, it’s evolved even more recently as we’ve done more extensive work in areas of the product, particularly with replication for example, where that alone has become valuable. Some of our customers want that almost exclusively.
So, we’re starting to rethink how we approach that pricing modeling, and looking at stratification of our product line. And I think what we will see going forward is some additional skews that allows to kind of address different levels in the market.
Michael Schwartz: Data persistence is a horizontal market – do you segment the market at all?
Ed Boyajian: Generally speaking, we don’t. And I think if you look at database, and just recognize a database is an infrastructure software, it is incredibly prolific across enterprise environments. I think if you look a little more closely in on database, if you look at the database like Postgres, which is a true, general purpose, transactional database, it has even broader horizontal applicability enterprise.
It is compared to some of the specialty database technologies that are in the market today. So, given that nature that it touches every application, it touches every department, it touches every developer, you know, we’re mindful that our opportunity is equally distributed, both in terms of within an organization, and then across industries and segments.
Now, having said that, we’ve seen centers massive for the business that have been prominent. And, so, we do some specialized marketing to those categories, and the ones that are most significant infotech, which is software and hardware technology. We have a biggest sub segment of our business government is much as a quarter of our business today, financial services, another really important vertical for the company, and then media, and telecom.
Those four segments would represent 90% of our business. We do Orion, some of our messaging and solutioning around those segments. And then split it other way, we are a global company with more than half of our business coming from outside the United States.
Interestingly, that split about 30% in APJ, 20% Amia, and the rest in North America. So, we also, in that context, do segmentation that is more geo-oriented.
Michael Schwartz: I’ve noticed that EnterpriseDB has really great partner network – what are the different types of partnerships that are important for EnterpriseDB?
Ed Boyajian: Yeah, it’s been an important part of our growth strategy. and I can only put it in context, we’ve had now 39 consecutive quarters of subscription revenue growth. And if you looked inside of that, that’s come largely because of a diverse network of go-to-market strategy than partnerships.
About 65% of our business comes through indirect channels, and so, we categorize that in the kind of classic form, distributor, reseller market, which is probably clearly the most prominent go- to-market route for us outside of North America.
We rely heavily on distributors and resellers, in both of Amia amd APJ, so, that’s a particularly mature and growing part of the business, in one that we prioritize.
Beyond that, we look at a network of partners who provide some other form of value out. We may try to call them, coin them OEM – they are not literally OEM partners, in one form or another. Or technology partners that bring EDB products to market, somewhat different than a distributor.
A notable one that we just announced this past year, late in the year, was our partnership with IBM, in close alignment with their data and AI teams, and where we bring our products to market. In close alignment with IBM Solutions, particularly around their cloud pack for data, for example.
So, that’s an example of a prominent partner that brings us to market. We see other examples of that, where we’ve developed strong alliances.
I think another notable one is Infosys, where, again, particularly in their new application development practice bring EDB and Postgres to their clients, as they do, to work there.
Michael Schwartz: Over the years, have you developed any rules of thumb, as to which partnerships to develop?
Ed Boyajian: I think there are partners who have a clear vision and agenda that relates to open source, and then, to go a step further, have gone beyond that and really defined Postgres as a part of that strategy.
Those, not surprisingly, are high-priority partner targets for us. And that actually exists across the continuum we’ve seen in the distributor and reseller world. We’ve seen partnerships evolve and emerge that were specifically focused with a partner specifically focused on bringing Postgres solution to market.
Even our recent partnership with IBM, for example, or our partnership with Infosys, our partnership that we announced last year with Ali Baba – all were centered on their vision to bring a Postgres solution to market.
Michael Schwartz: Building a global sales organization is a huge challenge – how did you go about transforming the sales organization when you joined EnterpriseDB?
Ed Boyajian: It was a really interesting challenge because I had just come from Red Hat, where when I joined Red Hat, the company was maybe 50 million, and I ran the North America business at the time of my departure, which was approaching a 250-300-million-dollar business. So, I had the opportunity to live through an extraordinary amount of growth in a sales model around open source.
When I came to EDB, the company was tiny by comparison, and there wasn’t a firmly establish sales pattern. So, one of the things that we started to do was, really focus on being efficient in acquiring customers. And I think it’s an important distinction, especially for this audience, rather than plow big money into what I think of is high-end enterprise raps, we started with inside sales motion that toggled around software downloads and incrementing sales spend at a relatively small rate, until we got to higher levels of a value with customers. And really built from what I consider really basic selling model out.
And that proved to be incredibly powerful because we got quite good at addressing customers. Frankly, in the way that I think they prefer inner set companies nowadays.
Modern Sales Strategy
Michael Schwartz: The enterprise software markets changed a lot in the last 15 years – can you talk about how you think vendors of open-source software should adjust their sales strategy for 2020?
Ed Boyajian: I think you have to look first at what’s changed in the landscape of IT consumption, and how buyers are forming inside companies. If you look in a little closer, first I think the commoditization of compute has allowed users across the enterprise to emerge, and you couple that with the way development is happening now more outside of IT, and more in the business unit.
The pattern of adoption for technologies has changed, its not centered on IT. So, I think the old strategies that we used to use in growing a global sales organization, to bring in relatively expensive high-end sellers, to engage with a relatively finite number of buyers has fundamentally changed.
The other thing that’s changed alongside that is, customers and users consume a tremendous amount of information on their journey. In fact, they’re heavily in self-service mode in their learning about technology or company.
And, here, again, I think an easy mistake to make in a go-to-market model, in an inefficient, go-to-market model is to think that you need to fill that blank in with staff, or we might think of it as high-end salespeople, but rather to build the systems that allow users and prospective customers to self-serve as far as possible, which is their preferred method of engagement nowadays.
So, my view of that has changed radically over the course of the past 20 years, in terms of how to build and structure the right kind of selling motions and sales organization.
Impact Of Cloud Computing
Michael Schwartz: As you know, enterprises are moving to the Cloud for many services, but as an enterprise software vendor, on-premise, or hybrid cloud, is still critical for growth. Is the move to the cloud accelerating? Has it peaked? And any advice for open-source software vendors on how to align with this trend?
Ed Boyajian: Well, I would say that the move to Cloud is accelerating. It’s hard to argue that, but I think, at least from what we see now, companies are starting to kind of settle in on enterprise strategies for how they deployed technologies.
Cloud is another important deployment platform, just as traditional on-prem deployments are an important deployment platform. And I think it’s easy to get caught in this kind of notion that there will be no tech deployed in what we may think of today as traditional contacts – we just don’t see that as a reality.
I think the difference for companies that are building businesses to think about what are the key technology capabilities that you have to develop to enable your customers to deploy in any environment of their choice – that’s how we’re focusing on this change. And we can’t avoid the reality that 70%, 80% of our customers deploy in traditional on-prem environments.
Now, those environments are becoming more cloud-like in the way they’re being deployed on containers, certainly in virtualized environments.
But, at the end of the day, our customers’ view, and we with them, view Cloud as a another very important deployment platform, but it’s just another deployment platform.
Cloud Strip Mining
Michael Schwartz: Amazon and other mega clouds are offering RDBMS as a service, and many in the industry are understandably concerned about Mega Cloud providers moving up in the applications stack. As someone who’s experienced this firsthand, do you think it’s a good or bad for your business?
Ed Boyajian: Look, we have many of our customers deploy databases in the Cloud, so they have to separate maybe two things. I think it’s very valuable to have Cloud computing as a utility and as a deployment platform for enterprise customers that makes compute more accessible.
I think it gives a lot of businesses, a lot of flexibility that they can’t get in traditional deployments. In that context, I think that the deployment of databases in Cloud is healthy for enterprises.
I think this question of whether or not Amazon being prominent in offering as a prime vendor of those database services, it intersects that, but I think most broadly speaking, I think the availability of Cloud services is a good thing, for companies like EDB, and for other open-source companies.
Advice For Entrepreneurs
Michael Schwartz: Last question, any advice for new entrepreneurs who are launching a business around an open-source software product?
Ed Boyajian: I guess I’d give maybe a couple of thoughts. One is, make sure that what you’re thinking about is where you want to take your business that you have – a credible, commercial vision for what you intend to do with the technology.
And I think that starts with really being clear-minded about the needs that are being served, and defining that in the context of target markets, and being thoughtful about the kind of business model you designed to serve that market.
I think it’s easy in open-source technologies to, at some level, get caught up in the enthusiasm that goes along with the projects. But I think building a business around that takes surprisingly a remarkable amount of thoughtfulness and discipline.
Michael Schwartz: Ed, thank you so much for making time to talk to us today.
Ed Boyajian: It was my pleasure – thank you.
Michael Schwartz: And thank you to the EDB team for logistical support.
Michael Schwartz: Transcription and episode audio can be found on opensourceunderdog.com. Music from Broke For Free and Chris Zabriskie. Transcription by Marina Andjelkovic.
Now, as promised, a special announcement time: for those of you who’ve listened to all the podcast, you might have noticed a preponderance of male voices. In fact, the male to female guest ratio is 42:1. The only female guest we scheduled was Jamie Thompson, way back in episode 2.
So, this year, we’re going to hear from more women who are leading pure-play, open-source startups. After this episode in fact, for the rest of 2020, it’s going to be all women.
To start things off, we have Deborah Bryant, Senior Director of the Open Source Program Office at Red Hat.
I had heard her given a presentation at the Open Core Summit last year, and I’m still repeating some of those things that she said. We’re filling out the schedule. If you know of any great open-source business leaders who are women, please let us know.
You can tweet at us using @fosspodcast, or you can send me a message on LinkedIn. Just mention that you’re a listener, and I’ll be happy to accept your connection.
The next episode will be out towards the end of February, and then you can expect episodes about once a month this year.
So, thanks for listening, and best of luck with
your open-source business models in 2020.