Thoughtworks described platform thinking years ago in plain terms. Remove friction for delivery teams with high quality, self-service access to the basics, and invest in a foundation that makes experiments easy. The author also warns that a platform is not “build it and they will come”. You build the most valuable capabilities first, in small steps, with the people who will use them.
In the Netherlands we have a word for that balance: Rhineland. Teams are free to solve real problems their own way, and a thin layer of shared standards keeps everything connected. Most platform programmes I see fail on one side. Either a central team builds a cathedral nobody asked for, or every team builds its own shed. A good platform is a paved road that teams choose because it is faster, not because someone told them to.