tecosystems

Billion Dollar Open Source Businesses?

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

One of the frequently debated topics concerning open source companies, both at OSCON and more generally, is how they compare in economic terms – or don’t – with their closed source or proprietary counterparts. The most common manifestation of this is the simple question: will we see a billion dollar open source business? Not traditional vendors leveraging open source to make a billion dollars, but a business built solely on open source reach the billion dollar revenue milestone. Before we attempt to answer that using crude but bottom line metrics such as market capitalization – which is impossible in a majority of cases anyway, it’s necessary to understand the fundamental compromise that open source makes.

In return for the massive distribution advantages that accompany its low or no barrier to entry approach, open source often (but not always) concedes up front licensing costs. This is in most cases favorable for customers, as they are free to pay at what Simon calls the “point of value” rather than at project commencement when value is still unproven. The question is whether or not it’s favorable to the businesses that are built on this model. Answer that, and you’ll be a fair ways down the path towards answering the billion dollar question.

The numbers, on the surface, would indicate that the various economic models embraced by open source firms are not in fact favorable to the firms that embrace them. Closed source vendors are typically appalled at the conversion rates of users/customers at firms like JBoss (3%) and MySQL (around 1%, anecdotally, on higher volume) – and those firms are more or less the most popular in their respective product categories. Even with the understanding that these are volume rather than margin plays, there are many within the financial community that remain skeptical of the long term prospects for open source (even as VC’s pour money in in hopes of short term returns). With the exception of Novell, who just made the cut with 2005 revenues of just over a billion dollars (and only a portion of those come from open source), none of the businesses within the LAMP stack (Red Hat/Novell, Covalent, MySQL, Zend, etc) are assumed (most are not public) to have driven revenues of $1B. Contrast those numbers with the revenues from closed source providers such as IBM or Microsoft that measure dollars with double digits followed by a B, and you might conclude that open source is just a flash in the pan because of the economics.

Such an analysis would be superficial at best and irresponsible at worst, however, and you are neither. So you look a little deeper and see that while open source might not (yet) create immense, monolithic wealth, it does benefit customers by lowering pricing and increasing choice. Further, it seems illogical to believe that even if open source can lower certain software acqusition and operating costs, those dollar savings are not invested elsewhere. How many CIOs will go their board and say “I invested in Linux, JBoss, & MySQL and saved us x dollars – please lower my budget accordingly”? You might also see that open source allows vendors to ammortize a number of traditional development, quality assurance and marketing costs, across a wide pool of volunteer resources, lowering the dollars they need to operate (you should hear Alfresco’s Kevin Cochrane talk about the delta in saleperson costs – it’s eye opening).

At the conclusion of these debates it’s common practice to deliver one’s one perspective on the question, for whatever it may be worth, so here’s mine. As stated at OSCON, I believe both that open source has reached and in fact passed the tipping point and is here to stay, and that it has a bright future. At this point, however, I do not expect any of the major pure open source firms to challenge the billion dollar threshold any time soon. The numbers – those that are available, at least – simply don’t support such a conclusion. But neither do I consider that a metric for success. At RedMonk, for example, we make quite a bit less than the analysts that come to us looking for work from time to time. We’re not crying poor – we do just fine – but we have intentionally kept salary expectations and expenditures conservative because a.) it goes with the open source territory, and b.) we expect open source to ultimately have a similar impact on the business of analysis as it does on the business of software. Is that a lack of success, or building for the future? Tough call. Open source in many respects seems to underpin a future in which more people will make less, rather than less people making more. I know which I’d pick.

The lessons that I draw out then are these:

  1. Forget the B, Focus on the V:
    I’d love to be proven wrong and see one of the open source firms rocket past the billion mark, but I’m not going to hold my breath. The history of software, however, whether we’re talking Microsoft or Linux, would seem to indicate that the best approach to wealth creation lies in volume rather than margin. So in that at least, the open source approach is sound. Were I an open source business then, and I suppose in a way we are, I’d focus on growing volume, community, and mindshare rather than the rather abstract dollar metric. It’s good to aim high, as long as you’re pragmatic while you do it.

  2. Brace Yourself for Impact:
    Regardless of what you may think of the economics of open source, they are probably here to stay. Assuming that’s the case, it would probably behoove every software business to ask themselves one simple question: am I prepared? Am I structured to either leverage or weather the open source storm, because if not there are some rough days ahead.

Disclaimer: With the exception of IBM and at times Microsoft, none of the vendors mentioned is a client.

7 comments

  1. that’s all good but a little 20th century stephen. 😉

    its a solid analysis but i feel its important to note that ALL software companies are mixed source software companies now. show me a software company and i will show you a firm that is using, and likely contributing to OSS.

    the irony is that, as per brandon wichard’s musing, there are companies “built on OSS” such as Google, making tons of money. but they are not major code contributors to oss. o’reilly’s architectures of participation.

    we’re seeing a related bet between nick carr and chris wagner (or is it lee gomes?) about profit driven versus peer driven content sites.

    surely success will be defined by a mix of both? we shall see.

    the reality is that the software business is now a mixed model business. BEA would call it blended.

    Brace yourself for impact- you got that right.

  2. I agree that for pure open soruce companies reaching the billion dollar mark in revenue will be difficult. In reality it is difficult for any software company; remember BEA is just over $1 billion in revenue now. I think James is correct, the blended model will be the primary model going forward.

    What I don’t buy is that open source is about volume? This sounds too much like the dotcom days of chasing eye-balls. Sure it makes sense if you are in the volume business but maybe open source is really about enabling more specialized software, ie. the Long Tail.

    Maybe we should think of the ‘v’ as being value creation. Open soource provides a more effective model for creating value since it allows you to 1) specialize the value creation, 2) lower barriers to entry so that it is easier to reach a niche and 3) reap opportunity cost by not recreating commodity infrastrucutre.

    Just a thought…

  3. james: “its a solid analysis but i feel its important to note that ALL software companies are mixed source software companies now. show me a software company and i will show you a firm that is using, and likely contributing to OSS.”

    agreed. even Microsoft has been shipping BSD software for a long time. but while the closed v open distinction maintained above is obviously artificial, i’d maintain that there’s blended and there’s *really* blended. i would not, for example, lump Microsoft into the blended category because they are not poised to embrace open source to the degree that, say, IBM is.

    ultimately, the point of the above is certainly not to argue that blended is not the way forward. it is rather to point out that those that are not blended will have a difficult time growing to 1B in revenues.

    “the irony is that, as per brandon wichard’s musing, there are companies “built on OSS” such as Google, making tons of money. but they are not major code contributors to oss. o’reilly’s architectures of participation.”

    well, i don’t know that i’d agree with that. while i agree with those who express skepticism at dibona’s comments from OSCON saying that “our code is so specialized it wouldn’t have an audience” i’m not sure saying that GOOG doesn’t give back to OSS is accurate. by sponsoring SOC, they actually give back to a variety of important projects in important and meaningful ways. they’re also reported to have folks working in house on things like OO.o. so i don’t think that assertion is necessarily correct.

    Ian: interesting pushback. value rather than volume? i’m skeptical.

    let’s look at eclipse. eclipse provides a value, of course, but is and has the value been that far ahead of commercial or open source competitors? i don’t think so. i think Eclipse was way out front in volume, and amassed the largest community as a result which locks people in via the network effects.

    is MySQL in the position it is because it offers a great deal more value than, say, Postgres? JBoss more attractive to partners than Geronimo because it offers more value? PHP so much better than Python, Ruby et al?

    i don’t think so. i think the argument can be made that value begets volume – though the argument would have a lot of holes – but over the long term it seems as if success in the open source world is defined more by volume than value.

  4. Open source stuck south of the $1 billion barrier?

    Stephen O’Grady apparently spent last week at the wrong conference. He came away from OSCON wondering if open source companies can ever reach the $1 billion mark in sales. He writes:As stated at OSCON, I believe both that open source…

  5. Stephen,

    I just think stating the success of open source is all about volume is too simple and sounds too much like eye-balls. For an organization to be successful they have to know what is their core competency; open source makes this even more important, due to the low barriers to entry. Maybe it is not value but if you define volume as downloads then I just don’t see it being sustainable.

    I also believe the world is moving towards more specialized/vertically oriented software. In this world large volumes may not exists but you can still be successful. I actually believe open source is making this possible. At Eclipse we are seeing more and more specialized IDEs that may not have been possible before Eclipse and open source.

    Certainly there are some very good examples of volume creating the network effect, Eclipse and the other you mentioned being valid points. However, moving forward I don’t think you need to have the massive deployment like Eclipse to be considered a success.

  6. Ian: i think i may see the crux of the problem, which is the volume = downloads definition. as you and i and mike have discussed before, i’m tremendously skeptical of downloads, believing that they can easily under or over estimate actual production usage. i would instead define volume as real world usage. in the case of MySQL, we can point to any number of datapoints that validate usage: companies big (Google), companies small (RedMonk), standardization in related texts (Agile Web Development with Rails), inclusion in bundled offerings (Ubuntu’s LTS LAMP option), and a dozen other examples and then, perhaps, downloads.

    i would also push back on the eye balls comparison, simply because where eye balls – IMO, anyway – had no real economic basis, volume does. real world businesses such as JBoss and MySQL are proving that the economics are real, if different than what we are all accustomed to.

    you bring up a very valid point, however, when you remind us that volume is not the *only* path to ubiquity. it may be, as i assert, be the surest, but there will unquestionably be opportunities in niche areas of the market for products to survive and thrive. i tend to think that few if any of the specialized players have a realistic shot at the billion mark, but then i don’t regard that as the criteria for success anyhow 😉

  7. Billion dollar marks are reach by… a billion dollars in revenue. Volume doesn’t = revenue (see Apache/Eclipse/Mozilla/OpenSSL) and downloads don’t = revenue (see Sun), and value doesn’t equal revenue (see Xen/OpenSSL/SNORT/Eclipse). Only someone buying a license or paying for support/services = revenue.

    $B of licenses… easy (and the market already accepts this model), $B for support …harder but possible (there is acceptance within mid-large enterprises), $B for services … much much harder (services = people’s time, $1B is a lot of people and a lot of time)

    For someone to buy a license or support you have to offer enough value beyond just good, open code.

    To get to a billion dollars, you need to offer enough value to enough customers that you get volume. At a billion dollars with value and volume you’re basically becoming an industry standard. So yes, I think Linux distros are prime candidates – they address the entire server market. Open source CRM/ERP apps… smaller market, entrenched “standards” already there, and a difficult market to get volume. Maybe sometime out in the future, but not soon.

    So they all seem to play together – tough to say just volume alone will ever get you revenue (ahem… Apache/Eclipse/Mozilla/OpenSSL – how much did Apache/Eclipse/Mozilla/OpenSSL make last year? they had plenty of volume). And it’s harder to say just offering open sourced value will get you there (CentricCRM is a great CRM app… have you bought it?)

    Besides… no open source project will ever reach a billion dollars – Larry Ellison always buys them or their underlying technology before they can hit $500M 😉

Leave a Reply

Your email address will not be published. Required fields are marked *