From Sun Tzu to machine learning, having good data is more than half the battle. The Army's acquisition data domain promises to prove that once again.
by Maj. Mario Iglesias
In the year 500 B.C., the Chinese philosopher-general Sun Tzu stated, "If you know the enemy and know yourself, you need not fear the result of a hundred battles." Sun Tzu understood how properly using data allowed leaders to make critical decisions leading to victory or defeat. Fast-forward to the present day, and the importance of synthesizing data continues to grow in modern militaries. Gen. Mark A. Milley, then-chief of staff of the Army, highlighted that lesson from Sun Tzu when he said that the Army lacks "the ability to see ourselves."
He and other senior leaders throughout the Army understand that accessing, visualizing and leveraging data has become a mission-essential task. Milley's statement in November 2017 launched the Army Leader Dashboard, a system designed to enable senior Army leaders to see data on all aspects of the Army, from personnel to logistics to acquisitions. The ongoing development of the dashboard has highlighted holes in our current data map, one of the largest gaps being the data surrounding our acquisition programs.
All of defense acquisition is characterized by the constant gathering of data. Every bit of a program must be documented, from need statement to requirements to every step of development. Every program has reams of data. Yet it has never been collected and managed at the enterprise level in any automated or systemic way.
The Office of the Assistant Secretary of the Army for Acquisition, Logistics and Technology (OASA(ALT)), for example, has always required programs to provide data for milestone decisions and in response to requests for information made by leaders and decision-makers. The milestone decision authority will require updated cost estimates, test data and a detailed schedule before approving a program's advance to the next phase of its acquisition life cycle.
Such data is collected, analyzed and provided in an easy-to-understand manner for the milestone decision authority and other stakeholders so they can confidently assess that the program should continue development and fielding. However, the acquisition community has lacked common tools across the enterprise that can store and provide the data for the dashboard to ingest before or after these milestone events. This is a problem for acquisition leaders and resource managers who need programmatic data at all points of a program's life cycle. This clear gap is both a challenge and an opportunity for the acquisition community to finally develop the tools that will fill the data gap and allow current and future leaders to make better decisions.
BUILDING THE ACQUISITION DATA DOMAIN
Dr. Bruce D. Jette, the Army acquisition executive, has developed the framework for the Army's acquisition data domain, which will be how the Army identifies, collects, manages and analyzes data throughout all Army programs' life cycleswhat it will look like and how it will function. The acquisition data domain will collect and link data from a program's inception as an idea through its development, production, fielding, sustainment and demilitarization. The larger domain will then interlink these subdomains so that leaders can understand the impacts of accelerating or divesting capabilities that are being developed.
Without such data, the acquisition community will not begin to leverage advanced analytical tools, such as artificial intelligence or machine learning. To do so, it needs access to the structured data that makes up programs. Building the acquisition data domain will require significant shifts in the business processes and tools that are used for all aspects of program management.
Last August, the Office of the Deputy Assistant Secretary of the Army for Plans, Programs and Resources began a pilot program to build the business management portion of the acquisition data domain. It is currently piloting the Air Force-developed system known as Project Management Resource Tools (PMRT). The PMRT system has been in use for more than 20 years, and comprises multiple modules for managing and visualizing programmatic and financial data.
The benefits of using a tool that another service has developed are the "speed to market" and design maturity. PMRT is already approved to operate on the Army network and is vertically aligned with Office of the Secretary of Defense reporting requirements.
Many of the program offices are currently using Microsoft Excel to manage billion-dollar programs. PMRT will replace these inaccessible spreadsheets and automate the way program offices manage their finances. However, some program offices and program executive offices (PEOs) have tools for managing their financial data. ASA(ALT) will work with these program offices to incorporate data into PMRT when possible. Moving from no toolsor several sets of toolsto a single tool used Armywide will mark a major cultural shift.
DATA CULTURAL SHIFT SPANS ALL LEVELS
The data transformation within the acquisition community will succeed only if the community, at all levels, actively takes part in the cultural shift. It will require users in the program offices to change tools and business processes while leaders at the executive level learn how to leverage the data that is being collected.
At the program office level, individuals must change how they conduct business and use the tools that will capture data as they work. The data in the system will only be useful to decision-makers if it is timely and accurate. Changing the tools and processes will inevitably incur a transition cost, such as for training and business process re-engineering, but it will be imperative for the success of these efforts. As trust in the tools grows, the number of requests for information that inundate the program offices will be dramatically reduced.
At the staff level, PEO and ASA(ALT) staffs will need to learn how to use the new data tools to collect information quickly without interfering with the program offices' work. Additionally, the domain managers will need to identify and adopt tools that will encourage program offices to use them. They also generally must be easy to use at various levels.
At the executive level, senior leaders will need to receive their information from the tools on the system. All of their briefs and updates on programs and initiatives should be sourced from the various data tools that program offices and staffs are managing. Leadership also will need patience and understanding, as there will be a learning curve and problems associated with using the new data tools. Additionally, leaders must leverage the data populated in the standardized reports to reduce rework by the staff.
TRANSFORMING DATA CULTURE
People who have been in ASA(ALT) for a while may have seen other efforts to transform how the organization uses data start and fail. The recent sunsetting of the Product Manager for Acquisition Business is the latest casualty in a list of ASA(ALT) data missteps. It is fair to ask how this new effort will be different from previous failed attempts. The answer is, there are a number of significant factors that will separate this effort from others:
Senior leader support. Reforming how the acquisition community makes decisions based on data is a top priority for the current ASA(ALT). Jette has maintained a focus on improving the way the organization uses data. His involvement ensures that the acquisition data domain initiative will receive the resources and advocacy necessary for making a large organizational change. Previous efforts did not have this continued senior leader involvement.
Learning from successful transitions. ASA(ALT) can benefit greatly by learning from commercial companies that have made large-scale, successful transitions. Recently ASA(ALT) hired McKinsey & Co., an international consulting firm, to develop a road map for building an ASA(ALT) data team and a detailed plan for realizing the acquisition domain. McKinsey has successfully completed similar projects with leading financial and telecommunication companies.
Technology advances. Over the last decade, there have been significant capability advances in cloud computing and software for managing data. The previous efforts created tools that were clunky and operated unacceptably slowly on the network. Tools today have slick user interfaces, and their capabilities continue to increase.
ASA(ALT) lessons learned. ASA(ALT) has a wealth of institutional knowledge on previous data transformation attempts. The current data team is reviewing the earlier efforts to learn what was effective and what was ineffective. As a result, ASA(ALT) is taking steps to mitigate the known risks and leverage the experience of those who worked on the previous data transformations.
DEMOCRATIZATION OF DATA
Currently, the acquisition community collects and presents data for decision-makers only at key milestones. However, the development of an automated system will allow users at all levels to begin leveraging data throughout the acquisition enterprise to conduct their jobs more effectively. This concept, known as democratization of data, is practiced in parts of industry.
There will be appropriate limitations on who can access and edit data, based on roles within the organization, but there won't be limits on access to the tools themselves. Once users see how these tools can help them complete their jobs, they will become more invested in maintaining and learning how to use them. With sufficient tools, the Army could optimize its investments and programs to maximize lethality over the next decade.
Companies like Amazon and Google maintain a sizable advantage over their competitors by collecting and leveraging data better than their peers. Everyone within a company has access to the data they need, when they need itin other words, it's democratized. Industry has seen the benefits of data management, and continues to invest billions every year in information technology systems and analytical tools that identify opportunities to increase revenue and reduce risk.
Many of these organizations are migrating legacy systems to fast and efficient cloud-computing centers such as Microsoft Azure or Amazon Web Services. Once the data is centralized, companies are able to visualize it and apply analytical tools, allowing better, more efficient decisions. These companies have demonstrated that leveraging data is essential for competing and winning in today's marketplace; the same will be true on tomorrow's battlefields.
It has become apparent that the acquisition community needs to invest in better tools and systems in order to effectively coordinate modernization of the Army. Now is the time for ASA(ALT) to radically change how the culture manages the data and decisions that allow the Army to optimize modernization.
For more information, contact the author at email@example.com or 703-697-4320.
MAJ. MARIO IGLESIAS is the strategic data team lead in ASA(ALT)'s Strategic Initiatives Group at the Pentagon. He holds an MBA from Yale University and a B.S. in economics from the United States Military Academy at West Point. He is Level III certified in program management and Level II certified in contracting, and is a member of the Army Acquisition Corps.
This article is published in the Summer issue of Army AL&T magazine.
- Data is decisive
Data is decisive
Last Updated : 6/13/2019PENTAGON - ARMY Editor