The topic name is meant to be controversial and elicit discussion; the proposal here is not to completely get rid of Capability Maps but rather to assess the merits of the Org Map as a better means to drill down to actionable Business Architecture. Over the last couple years, I've seen multiple research papers declare the value in drilling several levels in Capability Mapping, where Roles and Processes are certainly part of the characterization (and therefore the two approaches aren't orthogonal). The problem I see is one of Level of Effort; the Capability Mapping appears useful if performed very quickly and then providing context for a cross-functional collaboration in the business that then executes improvements through a focus on cross-business roles and key integrated processes. More often, however, it appears that the Capability Mapping is a great use of energy and resources with little business payoff, because it is drilled down multiple levels across the whole business without that focus on finding the 'center of gravity'.
Now look on your own website and you will find a little artifact called the Org chart. It already exists, and usually it can be cross-linked to leadership bios and lower-level Org charts for each part of the business. What it generally lacks, to provide the type of rigor and value an architect cares about, is a taxonomy (at least number the darn thing with a hierarchical outline) and a way to draw relationships, interfaces, functional partnerships, etc. Modeling the latter hardly requires a whole new Capability Mapping construct, it just requires a more dynamic, linked set of artifacts demonstrating (and maintaining) those relationships tied to the exact same Org chart. With this set of maps and a Business Strategy in hand driving key priorities, you have the means to model the roles and metrics for a core integrated business process that will execute that strategy.
Please provide your experience, views, counter-points, and examples.