“The Future of Software Exists in Great Ideas”: An Interview With SugarCRM’s John Roberts

0
1100

Share on LinkedIn

In 2004, John Roberts, Clint Oram and Jacob Taylor founded SugarCRM, Inc., with the idea that enterprise applications shouldn’t be proprietary, that once you implement a system, you should be able to easily modify it to suit your business needs. In this edition of Inside Scoop, CRMGuru.com founder Bob Thompson talks to Roberts, the company’s chairman and CEO, about SugarCRM and the open-source space.

Bob Thompson

I’d like to welcome to our Inside Scoop program today John Roberts, the chairman, CEO and one of three cofounders of SugarCRM. SugarCRM has received quite a lot of attention in the industry, and I guess you might call it the “sugar buzz.” And I’m looking forward to hearing John share some of his insights about how the company was founded, what its value proposition is and where it’s going. So, without further ado, John, welcome to our program.

John Roberts

Thank you, Bob.

Bob Thompson

Could you start us off by just telling us a little bit about your background and your current responsibilities at SugarCRM?

John Roberts

I am cofounder, along with Clint Oram and Jacob Taylor, and chairman and CEO of SugarCRM, Inc. I’m also a cofounder of the Sugar Open Source Project.

Bob Thompson

And what was your background prior to SugarCRM?

John Roberts

I have coming up on 15 years of building 100 percent proprietary, enterprise-class solutions in CRM and e-commerce enterprise applications. I’ve been in the [Silicon] Valley over 11 years, worked for some very strong firms over the years. And really, the basis of SugarCRM was looking at a new way to develop software using a distributed open-source development paradigm, mostly just because the base is being frustrated with the proprietary model, and having the belief that the future of software really rests in great ideas and great engineering. I think the market today has really been dominated by great sales and marketing and very little innovation on the engineering side.

Bob Thompson

What was the rationale behind the name, SugarCRM?

John Roberts

Historically, these applications have very low utilization and have been pretty difficult to implement. I think you have different classes of users of enterprise business applications, whether it’s in human resources, accounting, supply chain, manufacturing, project management. The CRM space tends to be dominated by an extroverted individual, and I think the issue with the space. It’s just been ultra-serious applications that the individual end-users haven’t really identified with. So the name behind Sugar is really just more of being, I think, a little bit more real with the end user and—

Bob Thompson

Maybe a little more fun.

John Roberts

In some ways, yes, I think, providing you have an ultra-serious customer relationship management application. But it is, hopefully, a little bit more entertaining to use. And if it is more entertaining to use, then your sales, marketing and support professionals use it, and it actually does increase the efficiency and utilization of the software.

Bob Thompson

Well, business is still a serious matter, and I’m curious more from the business model point of view: What was it that you felt made it a good business to get into? I mean, free software or, at least, some free software is something I’m sure all customers would like. But you just recently raised, as I understand it, nearly $26 million in another round of VC funding. I’m sure they want a return on their investment. I’m sure you want to have a successful business. So how do the notions of “free” and a successful software company come together?

John Roberts

Well, you’re right. We’ve always said from Day 1 that we are a commercial open source software company. And the reason why we’ve always put “commercial” in front of “open source” is because what always attracted me to founding this company is that open source is a very transparent world. You’re writing source code in public. You’re open-source licensing the code that you write. You have—in some cases—a very verbal community, and everything is on the table at all times in terms of how the software is designed, the requirements, the distribution, everything.

So by putting “commercial” in front of “open source,” we wanted to make it clear that there was a time when Clint and Jacob and I did work for free, but that’s obviously something that you can’t do forever. SugarCRM isn’t 100 percent for free, because designing and writing these applications does cost money. I mean, we’re 80 employees.

Community of developers

Bob Thompson

But on the developer side, you’ve got this developer community out there. How many people do you have contributing, basically, their time for free to make the core code better and better over time?

John Roberts

If you look at the core of Sugar Open Source today, it’s a pretty broad application in terms of sales, marketing, customer service, full-sweep reporting, offline support from multiple data bases. That is something that we control very tightly, because you have to have an incredibly stable code base, in terms of the testing, the release cycles that go behind that. And we have a community of developers around the world that we worked with over the last two years.

Just like Linux. If you look at the core kernel of Linux, it’s actually a pretty small community of developers. If you look at the broader spectrum of Linux, it’s several thousand developers. SugarCRM operates pretty much the same way. Sugar Open Source, we’re pretty tight in terms of the release schedules, the way that we manage it, the engineering release schedules behind it. But we did build in very early on and plug in architecture into the core code base that allows any developer around the world to build all sorts of extensions and modifications to the core code base, implement their ideas and make them available for other people to use.

Bob Thompson

So it sounds like you’re trying to get the best of both worlds—the innovation and free code, I guess it boils down to—from this community of developers while delivering a stable product to business users.

John Roberts

That’s correct. I mean, we’re very careful. I think what we’ve learned from the whole scope issue with Linux is that maintaining a very tight history of the copyright is really important for the end users—not only us but also for our users and customers around the world. So we are very careful. At the same time, what I most like about this model, having lived it for the last two years, is just the amount of incredible ideas that developers around the world have—not just the ones that live in Silicon Valley—but all over the world, whether it’s in Israel or Australia or China. It doesn’t really matter.

And so, we’re creating a framework that allows folks to write and express their ideas. But at the same time, we’re doing it in such a way that it’s not all dependent upon our release cycles. And that’s really where Sugar Forge comes in and this whole modular—what we call the module loader, which we implemented actually in 3.5 about a year ago—that enables developers to implement their ideas and share them very quickly and get feedback on them. Language packs is a great idea. We’ve been translated into over 40 different languages around the world, based on developer work.

We have Sugar Open Source and SugarForce.org that is strictly open source license software. Everything is completely free there. Our Sugar Pro and Enterprise products, we handle that very differently, in terms of, “Is it a commercial product?” We do provide the full source code. We do not believe in proprietary software. So there is a mix, and I think we do a very good job of separating what we call church and state, being Sugar Open Source. And then there is Pro and Enterprise.

Bob Thompson

And the Open Source, that’s the trial software for your new customers in many respects, isn’t it?

John Roberts

No, absolutely not.

Bob Thompson

Wouldn’t customers naturally want to start with something free and then try it out and say, “I’d want something that a company is going to stand behind and support?

John Roberts

That’s a good question. I think a lot of people like to kind of perceive it that way, but the reality is, it’s not trial software at all. There is zero restriction on its use. There is zero restriction on its capabilities.

Bob Thompson

Oh, no, no., I’m sorry. Let me back up. I didn’t mean that it was only good for trial.

John Roberts

OK.

Bob Thompson

What I meant was it is a route to becoming a paying SugarCRM commercial open source customer.

John Roberts

Yes, you’re correct there, Bob. The reason why I said it is because some folks like to categorize it as that, but that’s just not the case.

You’re right. Sugar Open Source has been downloaded over 800,000 times, and with all the extensions, we’re looking at like 2 million downloads. And it is completely free. We introduced two months ago Sugar Network, which is basically a support offering. So if you would like us to provide full customer support to your open-source installation, we do do that. It’s like $100. It’s not very expensive. But in general, it’s kind of a mix. Sugar Open Source, if you look at the design of the application, is designed for under 20 users. And that can be a sales team, a marketing team, a customer support team, a project management team. That’s really the design focus of Sugar Open Source. And we work really hard to provide kind of the ultimate solution, usually above 20 users. And we do provide Sugar Network, which does provide support at a very low cost for Sugar Open Source. Above 20 users for larger organizations, they want to have a relationship. They want more advanced features.

Bob Thompson

So 20-plus users? Up to what point?

John Roberts

Usually, around 20 users seems to be where we’ve historically made the decision in target professional and enterprise. For sites running larger organizations, above 20, it could be 500 users. At that point, those organizations traditionally don’t start with Open Source.

Bob Thompson

Above 500?

John Roberts

They traditionally download Open Source and evaluate it from an initial kind of cursory path to see if this is a potential candidate. So in that context, it is kind of trial-ware, but they know from Day 1 they’re going to go with Pro or Enterprise, because it is a super set of the same functionality. But it subscribes to the same belief system, in terms of full source code, open development and environment that you just can’t get from a proprietary vendor.

But we have just over 800 customers in just six quarters, actually, of shipping our core products—which is incredible momentum, and we’re really grateful for that. And tons of those are five- and 10-user customers that have chosen Pro instead of just using Open Source. And a lot of those have been running Open Source for a year, a year and a half or even six months. And once they’ve really seen the value and they want more at that point, hopefully, we earn the right for a support business. So it does work in that context, as well.

Bob Thompson

It sounds like your target market is kind of 20 to 500 and you’re going to be running into salesforce.com, NetSuite, Microsoft, Sage—lots of companies in the mid-market. Maybe I’m jumping ahead just a little bit, but in this particular space, it’s highly fragmented. There’s lots of different solutions. There are some pretty big brands out there, obviously, and some smaller, less-known brands that have very good technology. So, in the final analysis, why are customers purchasing SugarCRM versus some of these other alternatives?

Competition

John Roberts

Who do we really compete against on a daily basis? Right, it’s absolutely salesforce.com. It’s absolutely. And NetSuite, not so much today. Maybe in the future more and more but less so today. salesforce, Microsoft Dynamics, Siebel, Oracle, the older PeopleSoft products—you’re right, those are our competitors today. I’ve been surprised and pleased with the amount of traction. With salesforce, alone, we’ve done 30 replacements of customer base, which is not something I expected so quickly. And it’s for a variety of reasons. I mean, it’s not always just one reason. Sometimes they just like our app better. Other times, they want to go back on-premise, because they want more control. Other times, they just think it’s a better value. And it’s for a number of different reasons. Sometimes, it’s large corporations. Sometimes, it’s smaller ones. It varies. It’s not one answer.

I think, also, if you look at the overall CRM market space, even in North America, there’s over 5 million potential users of CRM systems in North America, alone. Yet, even the biggest vendors only have customer bases in the 10,000 to 20,000 range. Even the largest ones. So I look at our customer bases. Actually, probably over 50, 60 percent of it are companies that were just non-consumers of these kind of proprietary solutions that were on the market prior to a commercial open-source solution such as SugarCRM.

Bob Thompson

And what about the hosted or on-demand trend? Companies like salesforce are betting on that heavily, exclusively. Right Now predominantly sells that way and deploys that way; NetSuite. And there are others: Oracle CRM On Demand, the former Siebel product.

John Roberts

Sure.

Bob Thompson

So when I talk with executives that are selling their product, their message is: This is the wave of the future. Everybody’s going to be buying on demand.

John Roberts

Yeah, yeah.

Bob Thompson

You guys offer a mix. Can you comment on where you think this is really going? Is it really toward on demand, in terms of what you’re seeing? What’s going on out there?

John Roberts

It’s interesting. I spoke at an on-demand conference yesterday, and the whole thing was that the future of the entire software industry is 100 percent on-demand. We love on-demand. On-demand makes perfect sense for the right requirements. If you have zero IT, if you just want a server that’s completely managed and you’re comfortable playing by someone else’s rules, it absolutely makes a lot of sense. I think our approach to on-demand is a little different. I think it’s more modern and represents the real future of on-demand architecture, which is more Linux-based systems that are supporting hundreds of databases versus one kind of monolithic database, multi-tenant approach, which you needed to do 10 years ago. But most modern architectures today are racks of servers. Everyone has their own database application. If you design your application right from Day 1, which is what we did, you can still have the full upgradeability as if it were multi-tenant. But what you get in return is full control of the application. You have full control of your own private database.

We also provide the flexibility, such that if you ever want to go back on-premise at any time, you can. A lot of times, folks will outgrow an on-demand instance, and they’ll want to go on-premise for some reason. Because, with all the marketing aside, you can never have the control from an on-demand environment that you’ll have on-premise. And on-premise does make a tremendous amount of sense. I like to say it’s kind of like movies. DVDs are great, but that doesn’t mean people have stopped going to movie theaters.

Bob Thompson

What’s happening with your particular solution, though? Are you seeing the uptake shifting toward an on-demand? Is there a flip?

John Roberts

Sure. Today we have about 30 percent of our customers out of 800 are on-demand, and two-thirds are on-premise.

Bob Thompson

Do you think that’s where it’s going to stay or do you see a shift?

John Roberts

I would say that’s probably realistic. Our vision for the future of software isn’t locking people into one particular deployment option. Our belief is that the future of software is really giving power and control to the user and letting them decide which deployment option makes the most sense for them. In certain cases, it’s on-demand. In certain cases, it’s on-premise. And in certain cases, it’s an appliance server, that we also have. The future of software really revolves around the development, the design, the origin of the software, how it’s developed. In the proprietary world, if it’s on-demand, the on-demand guys will say, “No, it’s all about just on-demand, but, by the way, we’re going to develop everything in a very proprietary black box environment.”

Bob Thompson

I think [salesforce.com CEO Marc] Benioff might argue with you. Not that I’m speaking for him, but he’s been promoting the AppExchange as kind of an on-demand ecosystem of sorts.

John Roberts

It is, but—

Bob Thompson

How is that different from your approach with these third-party developers?

John Roberts

It’s kind of like a way where you can go and you can build applications, but it’s still a 100-percent proprietary salesforce environment. But if you’re comfortable with their proprietary standard of how you can write applications, yeah, you can write applications for that platform. It’s kind of like the proprietary development application tool sets that have been built over time. Our approach is: Let’s write software in public. Let’s allow folks to scrutinize the source code, itself. Write it in a modern programming language from Day 1. Allow developers around the world, whether they pay you or not, to participate in your ecosystem.

You don’t have to pay money to SugarCRM, in order to use Sugar Open Source and contribute to the community, but you do to be an AppExchange member. You have to pay salesforce.com dollars to use a 100-percent proprietary system. So I think that there is really a difference. I think, if you look at Sugar CRM from Day 1, we’ve used a distributed development model. There’s a reason why we’re in 40 different languages and salesforce is in 10 or 11 or whatever they list on their web site. I think the future of software is really about freedom. It’s about developing software in open, in public, making your source code available, not trying to lock people into proprietary development.

Vendor backing

Bob Thompson

But doing that for business application software has been, as far as I can tell, relatively rare. Like Linux is an infrastructure. MySQL, PHP—very popular but infrastructure building. One of your competitors, I would imagine, at RightNow—Greg Gianforte—said in an interview some time ago, that open source is great for infrastructure: We love it. We’re going to use it. But we think that the application ought to be something that there’s a vendor behind.

John Roberts

It should be proprietary.

Bob Thompson

Well, he didn’t use that word, but I think that was basically what he was saying.

John Roberts

It’s true. It’s true. [He was saying] the application should be proprietary. Our belief is: No, it doesn’t need to be proprietary. It’s interesting. We were the first open-source applications vendor ever to be funded by a top-tier venture capitalist ever. And I remember in spring of 2004, when we were going down Sand Hill Road [in Menlo Park, California, and known for the preponderance of venture capitalist firms] and meeting with all the top VCs, they would always say to me, “So, you’re going to take on salesforce and Oracle, and you’re going to be using an open source applications module approach. But ‘open source’ only applies to operating systems and databases and programming languages.”

I think a lot of it is if you look at the enterprise applications space, it’s huge. It’s such an under-penetrated market, and I think it’s absolutely the reverse. In terms of most individuals, when they get management information systems degrees from universities, let’s say, just in North America, what they’re trained to do is modify business applications. They’re not trained to build application servers or build databases. They’re trained to customize and implement business applications. That’s what I meant.

So, if you look at the pool of developers in just North America alone, it’s probably 10 times the number of developers that can customize business applications versus one of the others. I think the argument you hear from proprietary vendors is: We’re not going to develop our software in open, because if we develop it and use a proprietary model with “open standards,” we can lock you into a proprietary code base. And the argument is: If I give you source code, you’re going to screw it up. Well, yes. But why should the vendor make that decision? Why can’t you allow the customer to make that decision? And our approach is: Yes, we do believe that source code—that software, itself—is a really good thing. The IT part.

Bob Thompson

But is it for every company? I wonder if you could comment on the characteristics of the companies that can effectively use this approach. Is it literally for everybody? Or do they need certain skill-sets or certain platforms or environments that make it a more effective solution for them?

John Roberts

Bob, I’m loving your question. I think the proprietary vendors love to try to make a case that you have to be a developer to use open-source technology. That’s what they want you to believe.

Bob Thompson

Just for the geeks. If you’ve got geeks, then it’s great for you. Otherwise, for normal people, you need a real application.

John Roberts

Right. A proprietary application. And we can never give you a source code, because you’ll screw it up. We didn’t incorporate SugarCRM until four months after we founded the project. And the reason is: In an open-source world, the focus is that you have to write software that is incredibly easy to download, to install and to deploy with zero training. Zero training. If the software is easy to install, deploy, customize and get up in production with zero training—and delivers an incredible amount of value—at that point, possibly, the project will blossom. So if you need to be an engineer, if it is technical, if it does these things, the project, itself, will not materialize. It will not gain momentum.

In fact, it’s actually the reverse. I challenge anyone to download 99 percent of the proprietary solutions and try to get them going without any training. That’s why they rely so heavily on their professional services. So I encourage you to download SugarCRM today. There’s multiple installers for Mac, for OSX9, for Linux, for Windows. There’s multiple installers for all these different platforms. It’s easy. It’s actually the reverse of the concept that you have to be a developer. That’s a fallacy that proprietary vendors like to promote. It’s just not true.

Bob Thompson

Well, as far as I can tell, you’ve had an amazing impact in the industry with very little marketing. It seems like that’s part of the open-source model: building the buzz and so forth. So congratulations on that. But you’ve got some challenges ahead to really take SugarCRM where you’d like the company to go and, I’m sure, where your investors would like it to go. What do you see as the single biggest challenge that you face moving forward?

John Roberts

Today, you’re right, we don’t do a lot of marketing. I think marketing and sales should be 15 or 20 percent of operation—not 50 to 70 percent of your operation, which is, I think, the norm of proprietary guys like the ones we’ve been talking about. You can just look at their income statements and you can see how much they spend on engineering, relative to how much they spend on sales and marketing. So we’re very fortunate. We’re grateful for the growth of the Sugar Open Source project in the community because it’s voluntary.

Our focus has really been to write great software. And as long as we focus on listening to the needs of our communities, providing the infrastructure around the open-source community and focusing on writing great software, that will propel the company. That is the No. 1 thing that we focus on: listening, creating the environment, writing software in open, allowing people to extend it in all sorts of different ways and participate back in the community without having to be a customer first, a paying customer. I’m really excited. Six hundred customers in five quarters. I mean, that’s just incredible. We are incredibly excited.

Right now, we have customers from the Fortune 50 down to tons of SMBs. So, having founded the company two years ago, if I would have said then that we’d be at 80 employees, having raised $26 million in capital and that SugarCRM would be used all over the world in such a short period of time—we’ve had, I think, a pretty amazing impact, in terms of the established proprietary market—wow! I’d say this business model absolutely works. It absolutely is the future of software. It is a more innovative software. It’s less costly. It’s a much wider ecosystem than you can ever get in a proprietary environment, whether it’s on-demand or on-premise. It doesn’t really matter. It’s just a deployment option.

Bob Thompson

John, thank you very much. We try to schedule our Inside Scoop programs with innovative companies in the CRM technology industry. And it’s obvious that you guys have had a tremendous impact. So congratulations on that. And best wishes for much success down the road. Thanks for spending some time with us on our program today.

John Roberts

Thank you, Bob. I really appreciate you having me on the show. Thank you.

John Roberts
SugarCRM Inc.
John Roberts, CEO and one of three founders of SugarCRM Inc., is a pioneer in the open-source enterprise application space. Previously, Roberts held product management and sales engineering positions at E.piphany, BroadVision, Baan/Aurum Software and IBM. Roberts has a bachelor of science degree in business from Virginia Commonwealth University.

ADD YOUR COMMENT

Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here