Write One Specification For One Behaviour
Generally, each Behaviour Specification should be focused on a single behaviour, purpose or business rule. By doing this, each behaviour becomes an independent building block of trusted, tested and verified functionality.
When each single-purpose Behaviour Specification is implemented as an automated test, the process of discovering the root cause of an issue is very quick — there is little analysis required because there is no ambiguity about which behaviour or functionality failed.
A "line of trust" forms out of the understanding that each behaviour is itself tested and verified. When writing each subsequent behaviour, be aware of and use this "line of trust" to your advantage. There is no need to duplicate or re-specify the functionality that is already specified in other Behaviour Specifications.