Anonymous: Can you build a successful services company based on OSS?

[jasoncwarner]

Asked anonymously

How can you build a business model for a services company that leverages open source software without raising capital or simply being wiped out by existing players with access to the same software packages? In other words, what would an emerging or industry-specific Cloudera or Red Hat look like before hitting their current scale, how might that look different if they started now, and most importantly what would make the business defensible & profitable at all?

[jasoncwarner]

Yes, absolutely, though the degree to "successful" should be moderated in this regard.

Some very famous started as mostly service organizations on top of OSS: Canonical, RedHat, Mirantis, Confluent, Datastax.

That said, I do not think the traditional services approach will work in 2020. If you look at Cloudera and RedHat today they are much more traditional software companies with offerings, packages, etc etc etc. And they have both gone way more general (Cloudera is 'the enterprise cloud data company', not Hadoop etc). In fact, that is the only way I think you can build a big business on services, is you have to solve a general problem and your solution still needs to have extreme anchor tenant hooks.

Look at Databricks as an example. They are a data analytics company build on OSS that started around hosting the OSS project. If DB stayed just OSS hosted Spark, they would have not have been an interesting company and definitely note an $8b valuation company with massive upside.

My suggestion would to figure out what your general, not specific, niche would be and figure out the plan there.