Control

Step 13 constitutes control activities in the PICS framework. At a minimum, control systems in a DFTS process involve monitoring of and feedback for variations in process and software product quality for taking necessary corrective actions (see Figure 21.1). But in a larger organizational context, a solitary control mechanism such as these would hardly be effective, especially if DFTS is meant to be a key element of an organization's overall competitive posture. Simons, in his landmark work Levers of Control, proposes a system of four control constructs to successfully implement a competitive strategy:^[4]

1. Belief systems are an explicit set of values, beliefs, and purposes as expressed in mission and vision statements, credos, and statements of purpose. Commitments to customers, employees, and other stakeholders have to be clearly stated and practiced for them to be credible. They provide stability, character, and strength to organizations and serve them especially well in turbulent times and times of crisis. Johnson & Johnson's famous credo, listed in Figure 21.3, has been an anchoring influence for the company. It has changed little since it was penned by Robert W. Johnson, Jr., in 1943. It places responsibility to stockholders last; doctors, nurses, patients, and their families, employees, and the community all precede it. In more recent times, corporate leaders such as Medtronic's Bill George have argued against the futility of maximizing shareholder value as a sustainable strategy. He states: "Motivating employees with a mission and a clear sense of purpose is the only way I know to deliver innovative products, superior service, and unsurpassed quality to customers over an extended period of time. Over time, an innovative product or a service will be copied by your competitors. Creating an organization of highly motivated people is extremely hard to duplicate."^[5] Emphasizing customer and employee interests above stockholder return as a sound corporate practice is a recent trend in corporations globally. The Indian IT software outsourcing provider Infosys Technologies, which makes a lot of money for its stockholders, does not even mention stockholder interests in its mission statement while including clients, employees, vendors, and society at large (see Figure 21.4). Belief systems invariably transcend particular initiatives or performance cycles. They direct the behavior of senior executives seeking new opportunities or facing strategic choices that may alter an organization's course. Although belief systems constitute the most powerful control mechanism, their true significance is probably least understood.

   Figure 21.3. The Johnson and Johnson Credo

   Courtesy of Johnson & Johnson. Reproduced with permission.

   

   Figure 21.4. Infosys's Belief System (http://www.infosys.com/about/vision_and_mission.asp)

   

2. Boundary systems are formally stated sets of rules, limits, and proscriptions tied to defined sanctions and credible punishments. Such control mechanisms consist of codes of conduct, strategic planning systems, asset acquisition systems, and operational guidelines. These are meant to help senior executives in their conduct on issues involving the organization's reputation. They also define strategic boundaries to protect resources in opportunity-seeking behavior.

3. Diagnostic control systems measure and monitor outcomes to correct deviations from preset standards of performance as expressed in budgets, best practices, standards, goals, and metrics. The preset standards are set by senior executives or are negotiated between senior executives and those who report to them. To be effective, these standards must be based on the collective understanding of the beneficiaries of the outcomes, such as customers and other stakeholders. In a DFTS context, measures such as CoSQ (see Chapter 4) and SQMS (see Chapter 3) monitor process and software product quality. These are fed back upstream for corrective measures, as shown in the PICS framework (see Figure 21.1).

4. Interactive control systems require regular and personal involvement of managers in the subordinates' decision activities. Such systems include project management systems, customer appraisal and review, and profit planning systems. Such systems ensure that managers remain engaged and that the discussion is based on real and validated data that constitutes the agenda of discussion between managers and subordinates: the essence of fact-based decision making.

Another set of control mechanisms, internal control systems, are not used by managers to control strategy.^[6] However, they are fundamental in ensuring the integrity of data used in strategic control systems. It is therefore essential that those responsible for the organization's internal control systems also be involved in safeguarding the integrity of DFTS process and quality data, especially CoSQ and other financial data (see Chapter 4).

Finally, the process has operational controls. These are accomplished through feedback control systems, which are an integral part of the PICS framework, and through project management, which has its own control mechanisms. Organization-wide control systems are beyond the scope of this book. We will discuss feedback control systems and project management in a DFTS context.

Step 13: Feedback Control Systems

This is a crucial element of the PICS framework and is the equivalent of the "check" phase of Deming's PDCA Cycle (Deming called it the Shewhart Cycle).^[7] If you fail here, the whole process becomes erroneous and out of control. It therefore must be designed to operate well. A reliable feedback control system provides appropriate measures of key variables in a time- and cost-effective manner for learning the state of the development process and for taking subsequent corrective and improvement measures. The variations are measured from product as well as process appraisal and involve both internal review and interaction and appraisal with customers, both internal and external. These constitute a major learning opportunity for improvement and may consist of a cycle of review meetings and corrective measures. Such a mechanism is effective if it is an ongoing activity of continuous learning and improvement year after year until the initiative is effectively internalized as part of an organization's culture.

Case Study 21.1 illustrates GE's Operating System, which has proven to be a powerful tool for continual appraisal and enrichment. It transforms initiatives across several dozen businesses rapidly such that all the initiatives become operational across the company within one month of launch and produce positive financial results within their first cycle. GE's various initiatives undergo successive cycles of year-round review and enrichment through its Operating System. By 2002, Globalization had been enriched through more than a dozen cycles, Six Sigma had gone through five cycles, Product Services had gone through six, and e-Business had gone through three.^[8]

We propose a DFTS feedback review framework as part of the control mechanism, as discussed next.

Elements of DFTS Feedback Control System

Using multiple performance measures, financial and operational, was proposed by Kaplan and Norton in "The Balanced Scorecard."^[9] The DFTS feedback control system, shown in Figure 21.5, must be designed to measure variables that are important, measurable, and actionable. They must provide visibility of process and products meeting five sets of objectives (as you will see, they are interrelated): learning and teaching capabilities, CoSQ capabilities, customer satisfaction, process robustness, and improvement opportunities Here are some important questions they help address:

• Learning and teaching capabilities: Do we understand the DFTS process well and feel comfortable using it? Are we developing a required internal teaching infrastructure and resources to improve and expand it? If not, why not? What can be done to improve our learning and teaching capabilities?

  Figure 21.5. Elements of Feedback Control Systems

  

• CoSQ capabilities: Are we measuring the cost of poor quality accurately throughout the software development cycle? (See Figure 2.6.) Have we developed capabilities to use such information correctly? If not, why not? How can we improve our CoSQ capabilities?

  Figure 21.6. The Yearly Cycle of GE's Operating System

  Reproduced with permission from http://www.ge.com/en/company/companyinfo/operating_system/main.html.

  

• Customer satisfaction: Have we specifically met the customer's cost, quality, and delivery schedule, as measured by relevant metrics? Have we exceeded them or fallen short? If we haven't succeeded, why? What can we do to improve customer satisfaction?

• Process robustness: How capable is our DFTS process, as determined by a process feedback loop, in meeting stated and unstated customer requirements, as determined by CAR? What opportunities have been identified to improve our use of various tools, techniques, and methodologies that constitute DFTS technology? What steps should we take to improve the process capability before integrating the revamped DFTS process and expanding it further? How can we improve our DFTS process capabilities?

• Improvement opportunities: What are further opportunities for improvements, identified from internal idea generation and emergence of best practices, and as measured by the gap between the customer needs and process performance? (Process capability improves as the gap diminishes; this also implies less room for improvement.)

Establishing a Feedback Control System

Control systems essentially provide visibility on cost, quality, and other measures of performance. They often become sociopolitical in nature because the people involved may feel insecure or perceive a loss of power as a result of such visibility of their performance. A feedback system therefore must have the support of the CEO and top management to be viable. Another problem with such systems is that although they have been in use for all kinds of control purposes, they are not always perceived and designed well. The system should be developed by a team assigned by the DFTS steering committee. For any control system of this nature to work, it must be customized. Establishing the DFTS feedback control system involves the following steps:

1.	Obtain management commitment and support.
2.	Obtain the cooperation of users and other stakeholders.
3.	Establish a development team.
4.	Identify control system requirements and measures (see Figure 21.5).
5.	Determine sources of data.
6.	Establish procedures to collect data.
7.	Design control system reports.
8.	Collect data and conduct control runs.
9.	Eliminate bugs from the system.
10.	Implement, monitor, and improve.
11.	Expand the system.

DFTS Feedback Review Framework

A well-structured review system should be more than just a control mechanism. It should enrich and energize the initiative across the organization by providing required measures to improve the product and the development process. These measures let an organization improve and upgrade the system continually until the system has been perfected, integrated, and expanded. For that to happen, the CEO must drive the review process, involving the organization as well as its customers. It must be made a big deal and its success a defining undertaking for the business. The control system will be effective only if it is a hands-on, year-round process involving key persons within the organization as well as customers.

The measures needed at different phases of the DFTS process must be correctly identified. The control system should be integrated with the DFTS launch and its subsequent appraisal throughout the year. Any such new initiative needs the organization's CEO to be its champion for several years (see Case Study 21.1).

Case Study 21.1: GE's Operating System for Continual Learning and Enrichment^[8]

General Electric's Operating System is led by its CEO and provides a vehicle to launch a new initiative and enrich it through several yearly Operating System cycles (see Figure 21.6).^[12] We do not recommend that an organization blindly copy this system, because it is unique and an integral part of GE's culture. It is best that you learn from it and adapt it to your needs and context, rather than trying to copy it.

What Exactly Is GE's Operating System?

The Operating System is a review and learning mechanism and is "GE's learning culture in action." It is a year-round series of intense learning sessions in which CEOs, role models, and initiative champions from GE, as well as outside companies, meet and share intellectual capital.

The central focus is always on sharing, and putting into action, the best ideas and practices from across the company and around the world.

How Does It Work?

Meetings take place year-round, in an endless process of enrichment. Learning builds from previous meetings, expanding the scope and increasing the momentum of GE's company-wide initiatives.

Driven by the company's valuestrust, informality, simplicity, boundaryless behavior, and a love of changethe Operating System allows GE businesses to reach speeds and performance levels unachievable on their own.

What Are the Results?

The Operating System translates ideas into action across three dozen businesses so rapidly that all the initiatives became operational across the company within one month of launch and have always produced positive financial results within their first cycle.

Globalization has been enriched through more than a dozen cycles. Six Sigma is in its fifth cycle, Product Services is in its sixth, and e-Business is in its third.

The Operating System is closely linked to GE's other management tools and systems, such as its Social Architecture, Work-Out, Boundarylessness, and GE Values (see Case Study 20.2 in Chapter 20). All these are an integral part of the GE management system and produce a uniquely synergistic effect.

Table 21.3 lists the key elements of the yearly cycle spread over four quarters.

Table 21.3. Key Elements of GE's Operating System

First Quarter
600 global GE business leaders gather each January to evaluate last year's performance and to determine the coming year's key initiative(s). Subsequent events in GE are then centered on making these initiatives happen.
Month	Main Task	Who Participates	Subtasks/Activities
January	Operating managers meeting	600 leaders	Initiative launch Case for new initiative Outside company initiative experience One-year stretch targets Role model presentations Relaunch of current initiatives
February			Intense energizing of initiatives across businesses
March	Corporate Executive Council	35 business and senior corporate leaders	Early learning? Customer reaction? Initiative resources sufficient? Business Management Course (BMC) recommendations
Second Quarter
People leading and practicing the year's key initiatives present their progress to GE leadership. At the same time, an anonymous CEO survey is sent to 11,000 GE employees for candid feedback on the initiatives' progress..
Month	Main Task	Who Participates	Subtasks/Activities
April	Anonymous online CEO survey	11,000 employees	Do you "feel" the initiative yet? Do customers feel it? Sufficient resources to execute? Messages clear and credible?
May	Leadership performance reviews at business locations	All business staffs	Initiative leadership review Level of commitment/quality of talent on initiatives Differentiation (20%/70%/10%) Promote/reward/remove
June	Corporate Executive Council	35 business and senior corporate leaders	Initiative best practices Review of initiative leadership Customer impact BMC recommendations
Third Quarter
Role models and heroes from across GE (and across the world) present to the 150 corporate officers at an annual meetinga platform where each business can measure its progress against the best of its peers.
Month	Main Task	Who Participates	Subtasks/Activities
July	Session I: A three-year strategy		Economic/competitive environment General earnings outlook Initiatives update/strategy Initiative resource requirements
August			Informal idea exchanges at corporate and businesses
September	Corporate Executive Council	35 business and senior corporate leaders	BMC recommendations Clear role models identified Outside company best practices presented Initiative best practices (all businesses) Customer impact of initiatives
Fourth Quarter
As one cycle ends, another begins. Budgets, personnel, and business plans are finalized for the coming year. Continued review and analysis ensure that the right people are in the right jobs for long-term success.
Month	Main Task	Who Participates	Subtasks/Activities
October	Corporate officers meeting	150 officers	Next-year operating plan focus Role models present initiative successes Executive Development Course (EDC) recommendations All-business dialog: What have we learned?
November	Operating plans presented	All business leaders	Initiatives stretch targets Individual business operating plans Economic outlook
December	Corporate Executive Council	35 business and senior corporate leaders	Set agenda for January's operating managers meeting Individual business initiative highlights BMC recommendations

Each initiative goes through several yearly cycles of Operating System review and learning until it is embedded in the company's DNA.

Project Management

Project management competence is an important skill set for a successful DFTS implementation. The Project Management Institute (http://www.pmi.org) offers a Project Management Professional (PMP) certification program that covers the following body of knowledge:^[10], ^[11]

• The project management framework

• The standard for project management

• Project integration management

• Project scope management

• Project time management

• Project cost management

• Project quality management

• Project human resources management

• Project communication management

• Project risk management

• Project procurement management

Several brands of project management software are available, and a number of them do a fine job. However, they are not a panacea and cannot rescue ill-conceived projects. Such projects will result in poor quality and will cause time and cost overruns despite the excellent project management software. Most project management software delivers just fine in conjunction with DFTS technology.

Any detailed treatment of this topic is beyond the scope of this book. We recommend that project leaders (MBBs) develop such competence and earn PMP certification. Other members and champions must acquire project management competence as well. MBBs with project management skills could also be the internal training resource in this important area. Chapter 26 addresses the application of Quality Function Deployment (QFD) to software project management.