Search This Blog

Wednesday, July 22, 2015

THE USDA'S SYSTEM SNAFU

BRYCE ON SYSTEMS

- Another example of government waste and incompetence in building systems.

(Click for AUDIO VERSION)
To use this segment in a Radio broadcast or Podcast, send TIM a request.

Another major system snafu was recently reported by the press, "$444 Million Later, USDA Only Achieved 1.5 Percent of its Goal to Update IT System," (Washington Free Beacon - June 4, 2015). The project was sponsored by the United States Department of Agriculture (USDA) and reported to be two years late, $140 million over budget, achieved only 1.5% of its goals, and finally scrapped. Footing the bill for the disaster was, of course, the American taxpayer.

Named, the "Modernize and Innovate the Delivery of Agricultural Systems" (MIDAS), the system was intended to replace an earlier system, "Web Farm," which tracked information on farmers receiving aid from the USDA’s 31 programs.

According to a recent audit by the Office of Inspector General (OIG):

"In response to a longstanding need to modernize the delivery of farm programs, the United States Department of Agriculture's (USDA) Farm Service Agency (FSA) initiated a business enterprise solution called Modernize and Innovate the Delivery of Agricultural Systems (MIDAS). FSA reported to Congress in 2010 that $305 million would allow it to consolidate its 31 farm programs into MIDAS by the end of fiscal year (FY) 2012.

MIDAS is 2 years overdue and approximately $140 million over budget and has not delivered the promised enterprise solution. As of April 1, 2015, FSA had obligated over $444 million to this project and had retired only 1 of the 66 applications which were to be replaced by MIDAS. By 2022, the program is projected to have a total cost of nearly $824 million. In July 2014, Secretary Vilsack directed that future MIDAS development cease."

The OIG's biggest finding from their audit was, "MIDAS is Overdue and Over Budget Because of Ineffective Project Management and Oversight."

This is only partially correct. Understanding and applying the mechanics of project management is one thing, devising a road map to travel to your destination is another. From the audit, it was apparent the project lacked a systems design methodology to define the work breakdown structure, deliverables, and review points. Not surprising, developers were allowed to do whatever they wished with no direction. In an attempt to reign in control over the project, they kept reducing the scope to try and break it into more manageable pieces. The fact remains, you cannot effectively perform project management without an overall methodology. As we like to say, "Having a Project Management system without a methodology is like attaching a speedometer to an orange crate; it measures nothing."

Even though I know nothing about the system, other than what I have read in the audit, I can conclude the system suffered from such things as:

Poorly defined requirements - According to the audit, "The current project manager stated that, at that time, there were only high-level project requirements defined and those were never put into a detailed system requirement specifications document for the project." Without a proper set of requirements, what in the heck are you building? This means the developers wasted considerable time second-guessing what was needed. This, of course, lead to a plethora of "do-overs."

No analysis of the current system - Had developers truly understood the aging “Web Farm” system, they would have known the requirements they supported, the data collected and calculated within the system, its strengths and weaknesses, the business processes involved, and could have devised an effective transition plan. It seems rather obvious this was simply not done.

No test criteria or test plan - According to the OIG audit, there were numerous known defects in the new system, which caused users to become dissatisfied. Both the test criteria and test plan should have been devised earlier in the project so programmers understood how their programs should perform.

No documentation - Although it wasn't specifically mentioned in the audit, I suspect no documentation of any substance was ever devised. Without a set of blueprints, what in God's name are you trying to build? Or is this based solely on the builder's intuition? Obviously, documentation is needed for designing the product, as well as for maintenance and modification purposes later on.

Data redundancy - if there is no documentation, it is safe to assume data redundancy plagued the system. If the data is "dirty," the information produced will be inconsistent and unreliable.

Here is what I believe happened with the system: They took a software approach for designing MIDAS as opposed to a system approach. For example, they probably created a data base quickly, then tried to figure how to get data in and out of it. I would suspect the program source code was well written, probably using "Agile" techniques, but the fact remains none of it was designed to work in a concerted manner.

If you were to ask the MIDAS developers why they didn't concentrate on the important up-front planning and design phases, they would probably lament, "We do not have time to do things right"; Translation: "We have plenty of time to do things wrong." Had they spent more time in the initial design stages, there would have been less second-guessing, and the system would have likely come in on-time and within budget. The eternal question, to me, is why do people prefer to do things wrong? The truth is, today's systems developers do not get it, do not want to get it, and never will get it, which explains why we will never be able to build enterprise-wide systems again. Even if they did "get it," they wouldn't understand it as they simply do not care.

What occurred at the USDA is typical of the systems being built in this country today. For example, the highly touted Obamacare Health Care system was also well over budget, late, and cost taxpayers in excess of $400 million. This could have been done more competently and at greatly reduced expense.

Aside from not possessing the expertise to perform this work, developers simply do not want to, preferring to try and program their way to success. After watching this for forty years, I can tell you authoritatively, "It doesn't work."

Keep the Faith!

Note: All trademarks both marked and unmarked belong to their respective companies.

Tim Bryce is a writer and the Managing Director of M&JB Investment Company (M&JB) of Palm Harbor, Florida and has over 30 years of experience in the management consulting field. He can be reached at timb001@phmainstreet.com

For Tim's columns, see:   timbryce.com

Like the article? TELL A FRIEND.

Copyright © 2015 by Tim Bryce. All rights reserved.

NEXT UP:  WHAT EVER HAPPENED TO CIVICS? - Is it gone with the wind?

LAST TIME:  1976 REDS VERSUS 1927 YANKEES  - Which was the best team in Major League Baseball?

Listen to Tim on WJTN-AM (News Talk 1240) "The Town Square" with host John Siggins (Mon, Wed, Fri, 12:30-3:00pm Eastern); WZIG-FM (104.1) in Palm Harbor,FL; and KIT-AM 1280 in Yakima, Washington "The Morning News" with hosts Dave Ettl & Lance Tormey (weekdays. 6:00-9:00am Pacific). Or tune-in to Tim's channel on YouTube.