The Effective Methodology for System Requirement Analysis
By
Michael W. Klenc
November 19, 2001
MIS 488 Research
Paper
Joint Application Development/Design (JAD)
Ensuring the Success of Modern JAD
ABSTRACT
This research paper addresses the need for an effective methodology to gather vital information during the systems requirements phase of product development. Further, it focuses on how the technique of Joint Application Development (JAD) should be used in the requirements phase. In addition, the role of the facilitator of the JAD session, the incorporation of Group Support Systems technology, and the implementation of CASE tools during the JAD process are evaluated. Overall, this paper tries to define the most efficient methodology that businesses should use to solve problems.
JOINT APPLICATION DEVELOPMENT/DESIGN (JAD)
Purpose of Methodology
Collecting systems requirements is an inherently difficult process; hence, choosing the correct methodology to accomplish this task is very important. The traditional process consisting of one on one interviewing, questionnaires and observation, is too time consuming due to slow communication and long feedback time. (3) Furthermore, several articles present findings that stress the importance of an effective information gathering process. First, Bill Jennerich, in his website article, (2) lists some industry statistics:
These findings confirm that most errors found in the systems building process can be traced back to incorrect requirements or misinterpretation of the correct ones. (8) Later in the article, Mr. Jennerich lists the findings of a survey conducted by the Index Group of Cambridge Massachusetts in early 1990. The survey of 95 systems development directors found that:
In addition, Computerworld magazine (4) reported that:
Obviously, an inefficient and ineffective methodology can waste both time and money. Correcting erroneous requirements of a system in its operation stage is “at least a hundred times greater than correcting them at the requirements definition stage.” (8) Therefore, the challenge of systems development has become “building the right system by defining the right one” instead of “building the system right.” (8) As a result, businesses should implement JAD sessions with modern variations and strong leadership to eliminate most of these problems.
JAD History and Definition
Toby Crawford and Chuck Morris at IBM CANADA first developed the methodology, Joint Application Development, in 1977. They designed innovative methods to get IT professionals and end users to agree on requirement and design specifications. (1) JAD can be defined as a management process that groups end users and computer specialists to participate in a extremely focused workshop which allows information systems for the user to be integrated in a shorter time frame. (2,3) The key term in the definition is workshop. JAD features workshops that are “less frequent, more structured and more productive” (3) than meetings. Further, this workshop operates following a highly structured agenda with clear objectives including a method for resolving open issues that would normally hinder the design process. (2)
Despite being over twenty years old, JAD, with a few modern variations, is still the “best method for collecting requirements from users.” (4) In fact, many organizations utilize the JAD process for all types of decision making (not just for IT issues). CM Solutions, a consulting firm, advocates the use of JAD in problem solving sessions with their clients. A sample of their specific JAD template can be found at www.cmsolutions.com/tools/application_development/joint_application_design-jad.htm. (1) Additionally, many other consulting firms use some form of JAD to solve business problems. Even institutions are utilizing the JAD process. For example, the human resource department at the University of Texas at Austin actually provides detailed information regarding JAD on their web site: www.utexas.edu/hr/is/pubs/jad.htm. (5)
An effective JAD workshop should last 3-5 days, conducting 2-4 hour sessions and involving no more than 15 participants. (4) The participants include the sponsor, facilitator/leader, end users, developers, senior managers, outside experts, observers, and scribe/recordkeeper. Among all participants, the facilitator is the most critical member of the group. The group leader administers all activities, follows the agenda and tries to reach a consensus from the group.
JAD Philosophy and Purposes
There are several principles that comprise the JAD philosophy. Overall, the main purpose of the workshops is to create a common language. This common language helps all the participants communicate and understand each other’s needs so that IS can build systems that more effectively support the company’s higher-level information needs. (2) According to the University of Texas at Austin (5), the JAD philosophy incorporates four simple ideas:
JAD Advantages
In the JAD template of CM Solutions (1), they claim that after using the JAD sessions, clients should expect to receive three advantageous deliverables: product, people and process. Product refers to all the design documents (flow charts, data diagrams, systems specifications, etc.) and all documented decisions made by the group. Second, after participating in the session, employees (people) will be more skilled, more knowledgeable and more focused. Also, there should be an improved relationship among the group members. Last, the organization will have acquired the skills and knowledge (process) that will better enable them to manage their own sessions.
Furthermore, according to website articles, one by Alan Cline (4) and one by Bill Jennerich (2), several articles and case studies have concluded that the JAD process can yield tremendous benefits. Below is a list that combines the highlights of the findings:
Clearly, the JAD process is extremely powerful and effective. Businesses need to realize the importance of utilizing some form of JAD when embarking on new projects. However, businesses must also realize that no methodology is one hundred percent perfect. Therefore, the next section points out potential problems that can cause the JAD to fail.
Problems with JAD
The JAD approach encounters the same problems that traditional meetings do. Namely, a limited number of people can participate effectively before the meeting becomes inefficient. Most researchers believe that the maximum effective group size is five, which is much smaller than a typical JAD session. (6) Furthermore, only one person can speak at a time. According to authors Dennis, Hayes and Daniels in 1999(6), this fact leads to several problems. As one person is talking, other experts are blocked from contributing their ideas and information until it is their turn. During this period, these ideas might be forgotten or suppressed because they seem less relevant or less original at a later time. Basically, vital information may be excluded.
Also, some group members will dominate the topic, while others will remain silent. This domination and inequality of participation and influence can lead to poor-quality models that favor the dominating participants. (6) In particular, this phenomenon will occur in meetings where participants have different levels of status in the company, such as the case with JAD. The normal JAD workshop will include mid-level, as well as, senior managers. Members of upper management usually dominate the meeting and honest participation by subordinates is inhibited. (7)
Another problem that may cause a JAD session to fail is an ineffective facilitator. In fact, many JAD failures can be contributed directly to the facilitator. (9) Furthermore, to add to the concern, research has determined that most leaders and members of organizations are “woefully ill-prepared to meet the challenge of facilitating groups.”(10) Basically, businesses need to entrust their projects to capable and dependable leaders.
The next section tries to address the problems discussed here. First, Group Support Systems will be evaluated as a technique that can solve may of the group dynamic problems. Second, the role of the facilitator will be discussed.
ENSURING THE SUCCESS OF MODERN JAD
Implementation of Group Support Systems (GSS)
The aforementioned problems can be solved by the implementation of Group Support Systems in the JAD workshop. GSS are integrated computer and communication systems that support group work. (8) Specifically, group members use computers to interact and exchange ideas and information, in addition to discussing topics verbally. The goals of GSS actually address the problems listed in the previous section. GSS attempts to reduce the process loss associated with disorganized activity, member dominance, social pressure, inhibition of expression, and other difficulties associated in the group setting. (8) Furthermore, GSS provides at least three functions that may improve meetings: parallel communication, anonymity and group memory. (6)
First of all, parallel communication allows several participants to contribute information simultaneously. For instance in a two hour meeting with ten participants, group members have twelve minutes to contribute to the topic at hand. The remaining time of 108 minutes is spent not contributing. GSS allows all participants to type ideas simultaneously. Furthermore, the GSS software shares these ideas with everyone.
Second, anonymity may encourage participants, especially lower status participants, to contribute their ideas without the fear of ridicule from their superiors. Anonymity can also help alleviate the pressure of conforming to the ideas of the group. Essentially, anonymity will foster creativity and produce a more fruitful meeting.
Third, group memory is accomplished because all typed information is recorded electronically. All group members will have documented text of all past meetings. This memory may stop problems that occur when information is forgotten or misunderstood. Also, since all members can view the ideas of their fellow participants, they can mold initial ideas with their own, which will, in turn, bolster the creative process.
Lastly, research done by Dennis, Hayes, and Daniels in 1999, concluded that project managers perceived GSS to be as good as or better than the traditional JAD models in accurately defining the business process. Further, the results indicated that the GSS technique reduced the time required building models by about 75%. However, the authors admonish that using GSS may cause problems, which is discussed next.
Potential Pitfalls of GSS
The Dennis study encountered three fundamental issues. The first problem was user training. Participants need to be trained on the nuances of the GSS software. The ineffective use of the software will produce undesired results from the process. The authors recommend that participants regularly pause to ask questions and to confirm that they are correctly using the software.
The second problem involved consistency and integration. Essentially, different experts use different terminology, different semantic understandings of the process and different writing styles. Therefore, it may be difficult to incorporate all the information gathered during the meeting into one complete, efficient document.
The final issue was information overload. Traditional meetings can produce volumes of information by themselves. The addition of GSS greatly increases that amount of information. Therefore, participants may have a difficult time assimilating the tremendous amount of information.
Importance of the JAD Facilitator
As stated earlier, the facilitator is the most important member in the JAD session. One article wrote, “the two premier rules (for a facilitator) are: 1) the facilitator makes all the rules and 2) the facilitator can change the rules as required.”(12) Since the JAD facilitator is in charge of the entire process, the ideal candidate must have overall leadership qualities to effectively get results out a group meeting. The more important skills include being a good communicator, remaining neutral in decision-making, properly preparing for the meetings and knowing the business and relative IT techniques needed to accomplish the project. (9,13)
Communication may be the
most important factor, because the leader needs to effectively convey complex IT
techniques and solutions to end-users. (9) Listening skills are
equally important in communication. (13) JAD leaders need to
listen to discussions to ensure that the group focuses on the task at hand.
Second, neutrality is very important. The leader needs to guide the group in
order to reach consensus. Therefore, the leader can not have vested interest in
the outcome of the decisions and must maintain diplomacy when dealing with all
parties. (9). Third, good
preparation is the key to a successful JAD session. (2). It is imperative for the
facilitator to spend up to three weeks prior to the workshop to define all
objectives, deliverables, success factors, to select and prepare participants
and to prepare workshop activities. Last, knowledge of the business and of IT
techniques are critical attributes that the leader must have. The facilitator
can not effectively solve the business problems without this
knowledge.
The last section of this paper proposes the notion of adding CASE tools to the JAD process. Even though most CASE tools are primarily important in the design phase, consideration to these tools in the requirements phase has been increasingly popular. (8)
Use of Computer-aided Software Engineering (CASE) Tools
Computer-aided software engineering (CASE) refers to automated software tools used by systems analysts to develop information systems. These tools can be used to automate or support activities throughout the systems development process with the objective of increasing productivity and improving the overall quality of systems. (21) CASE tools can be classified as either upper, middle and lower. With regards to systems analysis, Upper CASE tools are primarily used because they “support models of enterprises, business environments, and planning.”(14) Furthermore, they are used in the application of various structured techniques including data modeling using entity relationship diagrams, process modeling using data flow diagrams, and structured design using structure charts. (8)
In terms of the JAD session, CASE tools are increasingly being used. In a research paper composed by Liou and Chen in 1994, they contend that “systems developers are better able to work closely with users in defining system requirements when it is possible to show them CASE tools-generated graphical models.”(8) Moreover, the article stipulates that CASE tools can help with the employment of GSS in the JAD sessions. Systems specifications, screen designs, and flow charts generated during the GSS-supported JAD sessions can be used as inputs to CASE tools for formal systems specifications, code generation and documentation. In conclusion, the introduction of CASE tools during the JAD requirement gathering stage of project development has effectively enhanced the design stage of the process.
CONCLUSION
The traditional methodology of Joint Application Development forms the foundation for effectively obtaining system requirements. The implementation of automated techniques such as Group Supported Systems will increase efficiency, while eliminating problems that arise in a group environment. Further, the process relies heavily on the abilities and skills of the product facilitator. Finally, the use of Upper CASE tools during the process will further aid in the efficiency of the entire system development.
REFERENCES