0/ Lots of founders have asked me to tweet storm my recent Open Source Paper into a quick 10 point hit list on what to look for when building an #opensource company. Here goes:
1/ Fulfill an operational pain point. This should be central to your company. Is this a nice to have for developers? Saving just a few hours of dev time? Probably not a pain point enough. When you get on the phone with your early users, do they say: "I couldn't survive w/o this?"
2/ Tech evangelizers from respected communities are critical in the early days. Developer behavior is the closest thing in software to consumer behavior. Word of mouth matters. If they like what you've built, they'll talk about it.
3/ Partner with enterprise users. When I mention "enterprise", early founders worry that they aren't ready to take on such customers. Use them as a focus group to understand their needs/requirements. If you aren't ready to take them on, its a signal you have more to build.
4/ Community, community, community. From managing in person meetups, to tracking slack channels and reddit threads - dialogue with customers and potential customers is critical...even if its negative feedback, you'd be surprised how good a user feels after he/she has been heard.
5/ Open source isn't as easy as SaaS to track, but here are some benchmarks I use as a north star: ReadMe Quality, MAUs, Github stars added/year, # of Fortune 500 customers, and # of contributors/maintainers.
6/ Know which business model you're tracking toward and where on the spectrum you are: pure services, open core and premium tooling, or open cloud. Some companies go straight to open cloud and don't get the traction they are looking for - start slow.
7/ Watch your usage patterns before you decide on a go-to-market strategy. Usage is a leading indicator for commercial success. If smaller biz's are using your product, lead with a low friction product. If larger biz's are interested, go straight to offering premium tooling.
8/ Have the commercialization conversation early and often. Does your team even want to commercialize? Not all OS projects were meant to become large businesses and you have to make sure you have the buy-in from your exec team and employees to chart new territory.
9/ You can build an OS business even if you aren't an OS project founder. There are lots of great OS projects that have no intention of commercializing (Elastic Search was repackaged Lucene, Confluent was built around Kafka.)
10/ Build an OS around a paradigm shift. In the last few years, the biggest OS companies were built around large technology paradigm shifts. If you see one coming - latch on quick (i.e. RedHat for Linux, Confluent around streaming, Hashicorp around multi-cloud.)
11/ These are not meant to be finite - just frameworks to help you brainstorm around an awesome OS project you've been thinking about, or to help you work through early business model issues in the path toward commercialization. Hope it helps!
12/ If you want to read at length about any of the above, feel free to read my more full-length paper using examples from @CRV's portfolio companies and my discussion with founders. / THE END
medium.com/crv-insights/t…
medium.com/crv-insights/t…
