The US Department of Defense's Acquisition, Technology, and Logistics (AT&L) organization won special recognition in Government in our SOA Case Study contest. Here is a snapshot of their story, created by excerpting the contest submission. The groupings and emphasis are mine.
Organization Background
The US Department of Defense (DoD) Acquisition, Technology, and Logistics (AT&L) organization has responsibility for management and oversight of Major Defense Acquisition Programs (MDAPs).
Business Scenario
Poor information visibility, manual report creation: Currently, DoD generally limits the collection, correlation, and dissemination of acquisition information to mandated reporting and specifically formatted products, such as the Selected Acquisition Reports (SARs) submitted annually to Congress or the Defense Acquisition Executive Summary (DAES) Reports. Oversight requirements from Congress and OSD are met primarily through manual data collection and the import of electronic data from the Services.
Acquisition program and status reporting processes for the Department are resource intensive, lack timeliness, and frequently are not authoritative. These limitations create an environment where the norm is reactive responses to existing problems versus proactive management to mitigate potential problems.
Improve decision-making: To address these issues and to provide effective and efficient delivery of warfighter capabilities, AT&L needed to achieve timely access to accurate, authoritative, and reliable information supporting acquisition oversight, accountability, and decision making throughout DoD. DoD leadership initiated a number of data and information sharing activities to address the information management gaps. One of these was SOA, as stated in 2007 AT&L Strategic Implementation Plan Goals.
Prior to SOA, data systems within DoD were implemented to electronically mimic pre-existing paper-based systems. They are better in that regard, but they are not noticeably different in type. SOA is reaching for a profoundly different data model: one in which data are simply, transparently available throughout the enterprise to whomever has a legitimate need for the data as soon as the data are developed.
ROI
Mission agility: To further prove the flexibility of this new data management model, the Air Force changed one of their authoritative data elements at the program management source for the B2-EHF program; business tool displays were refreshed, and the revised program data element was properly reflected in each display—available for use by the Department in less than a minute.
Better decision-making: The SOA infrastructure and related governance processes enabled authoritative information supporting acquisition decision making on $103B in total program value to be pulled from authoritative sources, as needed. With the existing processes, obtaining this same information may take several months, and the data may no longer be current when it is needed for decision making.
Project Organization
Center of Excellence. Governance Team.
Cross-service collaboration: The technical team was made up of technical resources from three sources: data providers (from Army, Air Force, Navy, and the Defense Acquisition Management Information Retrieval program office), the hosting partner (BTA), and software technologists.
Lessons
Data Governance Team: The governance established for the demonstration included leadership representing the functional interests and the technical interests of the effort, along with support from data source and SOA service consumer representatives. Mission: This governance body agreed to data definitions, identified authoritative sources for the data, and achieved data availability for on-demand access. Management: Centralized management is required for ongoing coordination of business processes and information requirements to avoid duplicate efforts and costs while encouraging reuse of services.
Security complications: Significant delays and challenges resulted from differing security standards among organizations, and initial lack of understanding of the maturity and configuration of data providers. Thorough analysis of security policies and practices and careful design are required for the next phase, which will include identity management, access management, Public Key Infrastructure (PKI), and the usage of Computer Access Cards (CACs).
SunGard won special recognition in Technology in our SOA Case Study contest. Here is a snapshot of their story, created by excerpting the contest submission. The groupings and emphasis are mine.
Company Background
SunGard Financial Systems delivers software and processing solutions to the Financial Services industry.
Business Scenario
Product transformation for agility: SunGard Financial Systems delivers software and processing solutions to the Financial Services industry through hundreds of products, many obtained through acquisition. These products have been built on disparate operating systems making integration very lengthy and difficult. These are the same challenges that their customers face with inflexible technology and business silos. SunGard is organized around different units, which serve unique business areas, such as investment management or brokerage and clearing. These units are known as business segments.
Customer Drivers: Many of their customers had embarked on SOA architectures themselves and wanted to find a trusted partner whose efforts complemented those initiatives. They also said they wanted to reduce the number of vendors they deal with and thus were looking to leverage more of what SunGard could offer.
SOA-ify Products: The goal with the SOA project (called Common Services Architecture or CSA) was to expose more components and business process logic and make it easier to consume them. CSA allows SunGard segments to catalog their software assets, produce reusable components and consume them in composite solutions. CSA establishes standards and fosters reuse through the sharing of design principles across products, and sharing of utilities and business services. The CSA is an evolution in the way that they develop new products and services.
The SOA project is addressing both the SunGard business problem of creating agility from siloed applications and the customers’ business problem of needing to change how the software mapped to their evolving business processes.
ROI
ROI is beginning to be recognized by the business with continued progress expected through this year and well into 2009 and beyond. The basic metrics of the ROI include:
1. Higher volume of solutions delivered, including greater exposure to previously hidden assets
2. Greater efficiencies of delivering solutions
3. Compatibility to integrate with customers’ SOA environment, resulting in higher business satisfaction.
Project Organization
SOA COE: The Infinity team manages the SOA Center of Excellence, which has a quality assurance team and certification and performance lab. They also provide centralized management of federated development communities across the global organization and the registration of assets and project management cataloging the decomposition of key products into agile components. Common architects represent each of the business segments and they review and vote on submissions to the SOA from various workgroups. Collaboration is fostered by community calls and active wikis to leverage each other’s work in real time.
Customer Advisory Board: To ensure that the project remains on track, they instituted a customer advisory board, whose membership is composed of enterprise architects and technology officers from top tier banking, investment services, energy and insurance organizations.
Lessons
They have learned through experience in the Infinity project that agility is realized far before significant reuse, (which does come but it requires a critical mass of component content before reuse becomes meaningful).
They also reaffirmed the importance that senior executive sponsorship has to an SOA initiative.
They gained an appreciation for some other key characteristics required for an SOA project of any size, including:
SOA & BPM: The interdependency between SOA and BPM has also become very clear to them. To achieve the benefits of BPM you need to provide the right level of services – to achieve business-focused SOA, BPM enables the identification and prioritization of which services should be built.
Penn National won special recognition in Insurance in our SOA Case Study contest. Here is a snapshot of their story, created by excerpting the contest submission. The groupings and emphasis are mine.
Company Background
Penn National Insurance is a regional property and casualty insurance carrier based in Harrisburg, Pennsylvania. Penn National Insurance writes approximately $550 million of annual premiums in personal and commercial lines of business combined.
Business Scenario
Losing market share, independent agent dissatisfaction: Over the past few years, the company has been losing market share in its personal lines of business due to competitor pricing strategies and dissatisfaction with existing systems among the independent agents that market the company’s products. At the same time, the company’s costs associated with processing new and renewal business in this market sector have consistently exceeded those of competitors.
Strategic business initiative: In 2006, Penn National Insurance made a strategic decision to commit corporate resources to reinvigorate its personal lines of business. Key elements of the plan included the implementation of predictive analytics to enhance pricing precision, and the replacement of existing systems to streamline processing and make it easier for agents to place business with Penn National Insurance.
To summarize, the company needed a new system, which enabled:
- Quality growth and market retention through increased pricing precision
- Expense reduction through lower processing costs
- Agency acceptance through ease of doing business
- Flexibility for adding new products, entering new markets, and providing a platform to absorb additional business in the event of an acquisition
- Increased data accessibility
- Faster time-to-market for new products
- Faster turn-around time for system changes
ROI
Pricing Precision: The primary objective of the project was to achieve better pricing precision. Less than twelve months after the project kick-off, a scorecard had been coded as a service and integrated with legacy systems. This allowed the company to move from a simplistic system that assigned new business among just three tiers, to a much more sophisticated system that evaluated numerous risk characteristics and then assigned the business to one of 25 new tiers, increasing pricing segmentation and pricing precision dramatically.
Increase agency acceptance: The second business objective identified was to increase agency acceptance of Penn National Insurance systems. The new system was available to Maryland agents in mid-April, 2008. During the first two months following implementation, the company saw a 65% increase in new business quotes from those same agents over the same time period for the previous year. Feedback from agents indicated that they were “extremely satisfied” with the new system, comparing it favorably to the systems of much larger national carriers.
Policy Processing Efficiency: The third business objective to be achieved by the project was expense reduction associated with policy processing. Using legacy systems, the company’s STP rate for policies processed without human intervention, stood at 18%. At three months post-implementation, the company is already experiencing an STP rate of 75%, where the new system has been deployed. When fully implemented for all states and all lines of business, the company expects to save $1.3 million annually in FTE processing costs.
Additional Cost Savings: Additional annual expense savings of $1.2 million are expected from using defined business rules that apply more discipline in the automatic ordering of third-party reports when compared to the previous process of manually ordering these same reports.
Further, the company anticipates an annual expense savings, in vendor maintenance costs and FTE development/testing time, of over $800,000, by eliminating the need to support two rating systems.
Overall, the company expects to achieve tangible annual cost savings of $3.3 million from the new system
Project Organization
Business & IT Collaboration: Business and IT project team members collaborated daily to define functionality details for specific modules targeted for reuse. (service design)
Another forum for collaboration existed in the company’s Data Stewardship Council (DSC). Comprised of decision makers from the Underwriting, Claims, Marketing, Finance, Actuarial, and IT Departments, the DSCs mission is corporate data governance.
Lessons
Business buy-in upfront: Time spent up-front gaining business support for the SOA project will be rewarded. There were many times on the project when they had to ask their business clients to think or work differently than they were accustomed to doing. Because their clients already understood the benefits of SOA, they readily accepted these new challenges.
Disciplined architectural governance and communication processes are essential to the success of a SOA project.