Leah Hickman's catalogue of roadmap failure modes, all traceable to the core problem in Roadmap (Feature/Project List) (dates turning ideas into commitments before validation):
Transitional roadmap (the bridge out): still lists features, but each one requires asking "why" and framing it against a problem/objective and a measure of success — a "working backward" step toward Outcome-Based Roadmap Reframing for teams not ready to drop feature lists outright.
The core problem isn't that roadmaps exist or that dates exist — it's that attaching a date to an unvalidated idea is the specific mechanical step that converts a harmless list of ideas into a trust-eroding promise. The antipattern is structural (the date-to-unvalidated-idea binding), not a matter of bad intent or sloppy attitude on the part of whoever wrote the roadmap.
A recurring antipattern: an executive's rank gets treated as a substitute for testing value. Idiodi's example: an executive says to build a Stripe integration; because it's the boss's idea, the team assumes it must be good and builds it without testing for value — then gets no signups. This is why "most roadmaps are challenging: value is assumed in a roadmap" rather than validated (Roadmaps as False Control, Outsourcing Strategy (Political Failure Mode)). The fix is the same risk-based trio work in Product Trio: value has to be tested, not inherited from org rank.