GE announced a few days ago that it was selling off its digital assets business, which to a large extent refects the failure of its industrial IoT platform, Predix. In 2016, GE predicted to make $15 billion in software sales by 2020 of which Predix was expected to deliver half of the revenues. Very clearly, Predix hasn’t taken off as expected.
I have been reading up on Predix and here are some of the lessons I gathered from my research:
- Deep pockets don’t guarantee success of a platform: Just because one is a huge manufacturer with dedicated customer base, it doesn’t guarantee success in building a platform of your own. Inspire of GE’s deep pockets and customer base, it did not guarantee Predix’s success.
- Supporting too many verticals: It’s difficult to build a software platform that works across many verticals. GE tried to be everything to everyone and built an all purpose platform for the wider industrial world, which was a diffused strategy.
- Wrong partner ecosystem: According to Parthasarthi V, your product needs to enable upmarket movement for your partners otherwise you will fail. If you see the business models of [GE] partners, none of them had opportunity to move up-market by leveraging Predix. GE selected a wrong partner ecosystem. There is no incentives for its partners to leverage Predix and move upmarket. The existing partners can only invest to a certain extent and they certainly did by training their resources, building Predix apps but it was not enough for Predix. GE couldn’t run fast because its partners could not run faster
- Building their own cloud data center: GE committed a strategic blunder to create its own Predix cloud data center to handle the data being produced by industrial asset. It took GE some time to realize that they were competing with cloud giants like Amazon, Microsoft and Google.
- IoT platform must be developer friendly: Predix platform is known not to be developer friendly. An IoT platform should provide future looking monetizing opportunities. It should be caring for the needs of professional developers. A good platform provides accuracy, findability and readability of developer content. Developers also need knowledgable support from the community.
- Not transitioning from product mindset to service mindset: According to Dima Tokar, co-founder and head of research at IoT analyst MachNation, GE failed to transition from a product-centric company to a services-centric company. He said that the transformation requires not only an evolution of human capital, but also an evolution in company culture, sales incentive structure and much more.