A Consultant's Casebook
The Project that Refused to Die
I thought I did pretty well getting one project on track. In the years before INTERNET, but after ARPANET, the United States Army Recruiting Command, USAREC, determined to implement its own distributed processing computer system. Rather than rely on the proven data communication protocols pioneered by ARPANET, the Recruiting Command solicited bids for the Army Recruiting and Accession Data System, ARADS.
PRIME the Michigan-based manufacturer of stand-alone mid-range computers was the low-bidder and won the contract. And although the Army Recruiting Command had its own MI (Management Information) department headed by William Bowe, the task of moving ARADS from concept to operational system was beyond the capabilities of his department. So, a contract was let to Ross Perot’s old outfit, EDS. The contract was open-ended! EDS was the sole contractor for ARADS until it worked!
Other consultants have all the luck! What a deal! Guaranteed lifetime employment if the system doesn't work. EDS was given working space in a separate building away from the prying eyes of the USAREC staff. Communication with the contractor was solely by work orders – USAREC staff wrote up a work order indicating what fix or software module needed implementation, and EDS responded by installing the fix.
Of the 7 modules called for in the design of the project, only one was working. The rest either didn’t exist, or didn’t work. The MIS department was under no pressure, they were managing the contract brilliantly as evidenced by their file cabinets full of work orders. EDS was under no contractual pressure, and could demonstrate its rapid response to MIS work orders. The fact that tens of millions of dollars and an operating budget of millions per year had not achieved results was conveniently ignored.
But the recruiters, and Army staff who needed to work with ARADS were very discouraged and complained until an evaluation of the project was ordered. The head of the Army Recruiting Command Planning and Analysis Department, Mr. Jüri Toomepuu, questioned whether the Army was getting its money's worth, whether ARADS would ever work as intended, and whether new, off-the-shelf software wouldn't be more appropriate. He asked me to review the system.
A review of the ARADS concept and proposal coupled with cursory interviews with those using the system suggested that Mr. Toomepuu was correct. This was clearly a sensitive piece of research that needed to be considered in its social and organizational context.
The first thing I did was begin getting our research train on track. With Mr. Toomepuu, we had an effective and powerful champion. I organized a research team that included recruiters, officers from the Planning And Analysis office, as well as non-commissioned officers from the recruiting command. We developed and analyzed questionnaire data administered to recruiters, guidance counselors at Enlistment Processing Stations, members of headquarters staff, and virtually everyone who had experience with the system. At that time, the only part of the system that worked reliably was the file transfer system from recruiters' offices to the regional ARADS computer, and that used Ward Christensen's public-domain xModem software. We wrote a scathing report concluding that "This project should die"
Our report received mixed reviews. The recruiters, guidance couselors, and those trying to work with ARADS were already on board. They agreed that the system was not fulfilling its promise. On the other hand, Bill Bowe of MIS viewed the evaluation as a threat to his management of ARADS, He had worked hard to make his department independent of the USAREC military command. MI functioned independently, and with an independent budget. His personnel simply ignored the results and refused to kill ARADS. The EDS team was certainly threatened by our review and also worked against our recommendations. General Wheeler, the USAREC commander had no personal stake in the system. He hadn't initiated the ARADS project, approved it, or championed it. If it didn't work, it wouldn't hurt him. On the other hand, any blame for killing ARADS would be laid at his doorstep.
So ARADS survives to this day - for over 10 years! This, despite the fact that in subsequent years, similar evaluations were done, each recommending that ARADS be scrapped as a waste of money. We couldn't have done our work better, but even a formidable juggernaut train can't cross every terrain.
All material is copyrighted! Not to be used without the written permission of Andres Inn. ©2003.
I appreciate comments and questions.
Send mail to Andres.Inn@mail.ee .