Architecture is more like a front window than a rear mirror. Architecture is a verb, not a noun.
EA (Enterprise Architecture) has to make a lot of tough choice for business and for itself, first of all, in face of today’s overwhelming business dynamic, what should architecture focus on: 'Today-AS-IS' or 'Future To-Be'? If EA focusing on the future, will it lead to 'fanciful states'? In which way, EA can deliver better value for the organizations?
Architect always included the TO-BE, and assessment of the AS-IS (where it doesn't fit the TO-BE), and a road map to move toward the TO-BE. It is hard to separate architecture from business strategy--it is the objectives and strategy that inform the architecture. It does make sense to focus the majority of EA efforts on the TO-BE, and that the TO-BE should be defined before the AS-IS.
Future states aren't developed in a vacuum or created as a fait accompli; they are developed with the key players from the business and IT. The road map informs people how they will get there from here. The road map is a commitment, a plan. As it's easy to blue-sky the future, becoming attached to alternatives that sound highly attractive, but which involve too much change from the current state (had it been documented) to be possible for the organization in question
The idealized end state assumes starting with a blank sheet of paper. The realistic state takes into consideration of today's current reality and produces some achievable target. The change equation needs a vision for the future. Doing architecture is to facilitate change, so why focus most of the effort trying to abstract the present which isn't going to generate any energy from anyone outside of the architecture community. As with everything in life, this too is a balancing act, balance between knowing where you've been, where you are, and where you want to go.
Identify the capability gaps in bridging ‘As Is’ & ‘To Be’: ‘To Be’ is not just the future state of architecture, but also about the future state of business, EA should provide input for business strategy which has already taken into account current state, SWOT and hence future directions. EA is then helping to identify key capability gaps and how to strengthen them or provide them (if non-existent currently). The gap from As-Is to To-Be plays a pivotal role in appropriate cost and timeline estimation, organization change management initiative requirements estimation and ultimately ensuring delivering value to the organization while retaining the learning and its differentiation from the past.
EA journey can be either horizontal or vertical, moving from bottom left corner to top right corner: Horizontal reflects attention to current, vertical reflects attention to future. One cannot get there without both. The journey will be different for different enterprises, it will even be different for the same enterprise at either a different time, or with different people, it will even be different for the same people done twice
Do only the amount of current state analysis that is needed in order to convincingly create a valid, achievable, realistic future state model. Understand where you have been and are to help avoid mistakes in where you want to go. However, too much obsessing over capturing As-Is configuration detail can waste time and resources rather than focusing on diagnosing the root cause of issues, then quickly moving on to define a set of integrated requirements.
What should the future window look like? Future needs to be further enough to overcome most organizational inertia but not too far out as to be unrealistic. After understanding the context and people dynamics, tailor your approach to ensure the right people are on board or sufficiently supportive to make it happen