Wednesday, June 10, 2009

Cycle Times

"The Perfect is the enemy of of the good."

"Staff work takes too long. I need to make a decision now."

"If they had all of the information, they would make better decisions."

"If I don't decide soon, I will be viewed as indecisive."

"There are implications of making a decision without having all the facts."

"We don't have time for that. Just use the best information we have and make a decision."

"We have to get context and content into the board room."

"Why does this have to be so complicated?"

If the statements above look familiar, then you've probably been exposed to the decision making challenge I call "Cycle Times." If timing is everything, then lack of timing is something that we should look at.

Fortunately, technology is allowing us to bridge the timing gap in some pretty spectacular ways, but before I talk about that, let's take a look at the cycle time problem in graphic format:
Everybody is right. Staff work (i.e. creating Enterprise Architecture models, analyzing them, combining them with other data sources like our budget system data, our technology repository data, business case information, etc) and producing a solid staff recommendation takes too long.

It is equally true that the current system we allow for making decisions has not evolved to allow for the kind of analysis necessary to ensure that decision makers are appropriately informed with respect to certain types of decisions. The complexity involved with managing a technology enterprise the size of the Department of Defense, and the method with which information information technology solutions have been deployed over the past 25-30 years make it nearly impossible for leadership to get the full picture without some degree of pre-meeting analysis.

The question is: how much analysis is enough analysis, and when is it time to just draw the line, make the decision and move on to the next problem?

The exact answer can be debated, but the evidence suggests that some degree of analysis - more than what we have now - is needed, and the time it takes to conduct a thorough analysis is too long. We are still struggling with most basic problems of visibility, interoperability and agility. Time to market is abysmal, and the GAO continually "encourages" the DoD to do things differently. We can move the lines.

The red swim lane above represents the as-is environment in many areas of the DoD. Decisions are made on relatively short cycles - say 30 days. An analysis team of engineers and/or architects, once notified they have a problem to analyze (which is usually after the leadership team knows about it), takes considerably longer than 30 days to produce models and scrub through available information to produce a meaningful recommendation.

The yellow swim lane represents what things might look like if we we to shorten the analysis time (perhaps through standardizing the due diligence process and establishing a core competency in this), and lengthening the time we allow for certain types of decisions. It still presumes that analysis starts to late and that the time to finish is still too long, but at least in this scenario, some usable (albeit incomplete) information is making it into the board room.

The green swim lane represents the ideal scenario. The analysis team is alerted in time to start preparing for a decision meeting. And the time it takes to get to necessary information is shortened. The end result is relevant, quality information available to decision makers at the point of decision making.


There are "authoritative" systems all over the DoD. The Defense Information Technology Repository (DITPR), the Select and Native Data Input System for Information Technology (SNAP-IT), and the Enterprise Architecture repositories are just a few. To my knowledge, the DoD doesn't have any consistent due diligence methodology for scrubbing all of these data sources and producing staff recommendations.

Any staff work done is usually done "off line" from the decision making meeting. Trusted agents who are close to the leadership summarize and advise their leaders before - and sometimes during the decision making meeting. Good "staffers" learn to anticipate questions or discussions that might come up and they hope to have as many answers as possible before the meeting begins.

Dashboards made, text is word-smithed, messages are "clarified," data is interpreted, visualizations of information are crafted - all with the intended purpose of getting relevant information into the board room when it is needed in order to make the best decision possible. For information technology investment decisions, this is not enough.

On the "soft" cultural side, Transformation agents need to work on expanding decision maker's expectation of what constitutes a good decision making cycle time. In some, but not all cases, a decision does not have to be made right away. Sometimes, the impact of making a decision right away is more costly than waiting for the analysis.

Transformation agents also need to work on the analysis end. Conducting an exhaustive analysis that is out of proportion for the decision that needs to be made must be unacceptable. Fit the analysis to the problem, then use technology to speed the answers into the board room.

On the "hard" or systems side, we have to ask ourselves if there is anything in our existing capability tool set that would allow us to hit all of our available "authoritative" data sources with just a few mouse clicks. Do we have anything in our tool box that would allow us to "mash" all of that authoritative information together and give us a view of the world in near-real time? We need that kind of speed and power to bring our analysis cycles in line with decision cycles.

I have personally witnessed technology delivering on its original promise. The Business Transformation Agency has figured out how to use technology (through Web Services) to tap dozens of "authoritative" data sources, mash the data together, and spit out reports in near-real-time. I have seen DITPR (Defense Information Technology Portfolio Repository) data laid up against SNAP-IT budget data, investment review dashboards, Enterprise Architecture, reports, and business case information with a few clicks of a mouse. Imagine the power an investment review board or an analysis team would have with that kind of information at their fingertips - exactly when they need it.

This information is available through the BTA to any DoD government employee with a Common Access Card (CAC). Smaller investment review boards throughout the Department of Defense can pull information from other areas in the Department of Defense for their review board or PfM projects. Someone sitting in an Army command can look at what's going on across the Department in the Navy or Air Force (and vice versa). High level boards and low level boards can pull from the same source and see what the others see. This is a PfM enabler on steroids.

No comments:

Post a Comment