Playbook
Platform Convergence
The playbook every other playbook on this site depends on: collapsing fragmented product silos onto a single, shared platform infrastructure.
The joint-planning gauntlet
Convergence fails when it's treated as an architecture diagram instead of a planning exercise. It succeeds when every product manager and engineering lead who owns a silo sits through the same joint-planning gauntlet and leaves with a shared roadmap.
The convergence sequence
- Map every silo's dependency surface first. You cannot converge what you haven't fully inventoried — undocumented integrations are where convergence timelines die.
- Run joint planning as one gauntlet, not sequential meetings. Put every silo's product and engineering leadership in the same planning cycle so trade-offs are negotiated once, in the open, not renegotiated per team.
- Design the shared layer for the laggard, not the leader. The platform's first version should absorb the least mature silo's requirements — it's the one most likely to resist migration otherwise.
- Migrate on a forcing function. A shared platform without a decommission date for the old systems becomes a permanent second system to maintain.
Why this is the foundation playbook
EBITDA defense, AI monetization, and NRR acceleration all assume a platform capable of supporting them. Convergence is the prerequisite work — the layer every other initiative is built on top of.