How might we help to ensure the success of information technology projects and the quality of information technology products?  Checklists of the critical factors for both project success and quality products are described in detail. 

The project checklist is based on a comprehensive review of the software literature for "early warning signs of project failure" and "critical factors for project success."  On the other hand, the quality checklist is based upon my synthesis of W. Edwards Deming's 14 Points and Joseph M. Juran's Trilogy.  Both checklists are organized, prioritized and weighted.  These checklists might be utilized, with or without their weights, to evaluate our planned, probable or actual performance and to identify how to improve our likelihood of success.  If the weights are added for factors that are accomplished, then this score might simulate the "probability" of success.  A partial weight should be given for a portion done, and success should be enhanced by higher-weight factors. 

Although we in information technology have talked about user needs for over three decades and quality products for almost two decades, we have not yet seriously confronted customer satisfaction.  If we are to achieve customer satisfaction before outsourcing overwhelms us, we need to significantly improve the success ratio on our most important projects and complex systems. 

In addition, this approach provides insight into the anatomy of success and happens to be a game anyone can play.  Attendees will be encouraged to utilize the checklists on their own projects and to suggest improvements.

After leaving Stanford with a PhD in Mathematical Statistics, Dr. Goodman spent 35 years working with information technology (aka data processing) in aerospace, petroleum and then county government.  He has been responsible for problem solving, planning, performance evaluation, management science, computer chargeback and computer capacity planning.  In addition, he has been involved in the Quality Movement since attending Ed Deming's very first 4-day seminar.

Dr. Goodman presented a very interesting talk on use of a checklist and use of surveys of both managers and customers to determine the important things to do in order to assure successful projects with high quality results.  The history of software projects in recent years has not been good.  Surveys have indicated that 34% fail and 50% are 200% over budget, leaving only 16% regarded as completely (in these terms) successful.  Something has to be done to improve this situation.

The work that Dr. Goodman has done stems from that of W. Edwards Deming and Joseph M. Juran.  Deming stipulated 14 quality points which came from a statistical and quality control background.  Juran listed 15 points that were more engineering- and product-related.  Dr. Goodman was able to map Deming's 14 points onto Juran's 15 and Goodman constructed an extra point from Deming's published views to match Juran's.  The mapped statements were different in wording and addressed each issue somewhat differently but were a very good match for overall meaning.  Dr. Goodman also presented some longer sets of detailed items that were given decimal rating values indicating their importance.  Weights of factors are initially obtained from experience and the literature.  Weights are improved using the customer's assessment of their importance.  A perfect score on the checklist would be 1.00.  The checklist follows the theme that "Success is a sum of its critical factors."

The lists are obtained by collecting both "early warning signs of project failure" and "critical factors for project success".  A major part of the work is identifying these to organize, prioritize and weight the individual items.  The checklists can be used to identify things that must be accomplished during a project before a it begins, and they can be used during a project to help determine current progress, and they can be used after the end to determine what happened and why.  In too many cases in the past, check lists and other types of software procedures have been aimed too much at the developing group and not enough at the customer.  It is important to get the customer's reactions early.  The views and priorities of the customer frequently vary from those of the developers, unless there is good communication and the views of the customer are taken into account.  If the customer is internal, failure to meet the requirements can lead to outsourcing.  

Dr. Goodman said that outsourcing is usually justified by claimed reductions in cost and by the availability of better outside technology.  However, it usually means that the internal customer is dissatisfied with results of (previous) projects and especially their usability in meeting customer needs.

After Dr. Goodman had gone over the generic checklist, he presented a more specific list for Data Processing Success.  He followed by presenting the results of a real checklist filled out by patients at a hospital.  He concluded by noting that scorecards are excellent devices for management to use in budgeting and quality control.

You may contact Dr. Goodman by email at agoodman@uci.edu.  This was the ninth meeting of the year, and was attended by 25 people.

Mike Walsh, LA ACM Secretary