The Qcon conference in San Francisco has always been one of my favorite conferences. Floyd is doing a great job of bringing an interesting blend of people from across the spectrum of the industry (Java, .Net, Ruby) into one place. He also brought some interesting speakers that you don’t normally see at this type of developers' conferences, such as the VC’s talk, which I found particularly interesting. This conference is a great environment to open your mind to other ideas and thoughts outside of your day-to-day realm. It took me a few days to let all the experiences from the various discussions in the conference sink in.
Obviously it is impossible to try to summarize three days worth of discussion in a single post, or even in a series of posts. I therefore picked out a few topics that I thought were the most interesting. I’ll start with the VC’s keynote speech.
Part I - Techie VCs Talk about Trends & Opportunities
In this keynote speech, Kevin Efrusy from Accel Partners and Salil Deshpande from BayPartners shared their successful experiences with open source companies such as SpringSource, Hyperic and Grails, and tried to draw a pattern for building a successful business model in the current market economy. Below are the main points that I took from their discussion.
OSS/SaaS/Cloud has a Common Driver
OSS/SaaS/Cloud reduce the barriers to entry to consume new technology. As a result of this, we are seeing a major shift in the technology selection process today compared with previous years. Technology is now being selected by those who are going to actually use it, rather than by the business managers. These users value simplicity, openness and productivity more then big brands. They are much more open to new technology as long as it serves their productivity needs. This shift in the decision making culture is also reflected in those companies' structure. It is now much more common to see senior management that is driven by a similar profile of technical leadership, rather than by business school graduates. Geva Perry gave an interesting explanation for this. In today’s world, innovation becomes key to the success or even survival of many companies. In such an era of innovation, technical leadership tends to have a better intuition for making the right choices that will make their product more successful then others.
Stephan the lead architect of Unibet, an online gambling company, provided an interesting insight during his presentation, on how he makes a technology choice:
- Open source software and open standards should always be the first choice.
- Avoid vendor lock-in. Software that is used should have a right-to-use license without any cost attached.
- Commercial, proprietary software needs to show exceptional business value (over free solutions) in order to be considered.
It's Not Just About Price
Unlike what most people think, the actual cost of an OSS/SaaS product can be similar to any commercial offering, or even more expensive if you start to measure the ROI. The core difference is the fact that with OSS or SaaS, you pay only when you get real value. You also get the choice to determine when you are willing to pay.
OSS/SaaS or Cloud = Cheap Marketing
Kevin made an interesting observation WRT to the business value of OSS/SaaS. If you take away the “religious” aspect of those destructive models, one of the main business values behind OSS/SaaS can be summed up as “cheap marketing”. You can get a quick channel to a large community that you would probably never have gotten if you are not on that side of the spectrum.
How to Monetize on the Success of an Open Source Product?
The general rule of thumb is to monetize for the things that are considered high value by your customers and not for things that are of low value, like development or tools etc.
Examples of high value features are features that are relevant for the production system but less relevant for development, such as:
- Deployment automation
- Support (SLA)
- Monitoring/Administration/Automation
- Security
Examples of low value features:
- Development tool
- Training
Charging on those low value items can be perfectly fine for seeding your company, but this is not scalable as a long term strategy. It is also not mutually exclusive, meaning that you could still have a training business along side your other source of business. The important thing is not to rely on training as the main source of revenue for your company growth.
How to Beat the Big Players
- Rely on one of the disruptive forces (OSS, SaaS, Cloud). Leverage the low marketing cost of a community-driven project to gain fast awareness (mostly through word of mouth).
- Start with small components (feature vs. platform) and grow slowly through the value chain. An exception to that example is JBoss - JBoss owes its success to the adoption of J2EE. It is therefore less likely that this model can repeat itself as there is nothing similar to J2EE on the horizon.
- Once you get to the right level of adoption, you need to start building value quickly to be able to monetize on the community. The right acceleration model is acquisition of other tools in that area.
- Focus first on adoption (at the expense of short term revenue), and monetize later. It is very likely that when you start to build your community, you wont have a clear answer on how and where the monetization will happen. The answer often comes somewhere down the road. It is very likely that it will involve a long trial and error experience until you figure out the right combination that will drive revenue out of your community.
Interesting examples in that regard are LinkedIn/Facebook, which are both now profitable and growing fairly fast. When they started they didn't really know what was going to be their main source of revenue. Google is another good example of that.
Main Hot Trends:
“Big data on the cloud” was marked as one of the “hot trends”. Unfortunately I haven't found my notes on the rest of the hot trends that were mentioned. I hope that either Kevin or Salil will comment on that directly.
