The Big Idea: How much should Dragon Systems, a Northern Virginia-based technology infrastructure company, bid to secure a contract from a large agency of the federal government? Bidding too aggressively (low) risked winning the contract at an unprofitable price. But there was a lot of competition for this contract and a less aggressive bid was unlikely to win. Dragon Systems needed to strike the right balance to maximize expected profitability.

The Scenario: A prominent agency of the U.S. federal government issued a Request for Solutions (RFS) for a high-profile, multi-year project in the agency’s historic office building in the District. The agency was seeking the design and installation of an integrated information technology network. Competing bids would be assessed for technical competency and price.

The building, a short walk from the White House, covered three city blocks and housed 4,000 employees. Because the various units of the agency operated independently, computer networks within the building were both physically and technically isolated from one another. With no shared information system, the units were forced to use the Internet to transmit data back and forth. Cabling and networking equipment were scattered throughout the building, snaking across open floors, sitting on desks, stacked in coat closets. Reassigning personnel often meant reinstalling cable at considerable expense.

The RFS sought firm fixed-price bids to address these problems. According to the bid request, the government estimated that the work would cost $9,042,000. Dragon System’s management suspected that this cost estimate had more to do with the budgetary constraints imposed by Congress than an actual estimate of the cost to do the work. The RFS specified that contractors would have to take care to preserve the historic nature and character of the 1.8-million-square-foot building (e.g., the roof’s terra cotta tiles) while working around the staff.

The uncertainties surrounding the building and what would be necessary to preserve it cast a long shadow over the bidding process. Dragon’s best estimate was that it would probably cost about $11 million to do the work. It did, however, believe there was the potential for follow-on work that was outside the scope of the currently proposed contract. The exact extent of this work was unclear.

Each year, Dragon submitted anywhere from 50 to 100 competitive bids ranging from $500,000 to millions of dollars. Its win rate approached 30 percent. Most contracts were bid based on cost and generally included a 5 to 15 percent profit. Having successfully completed more than 60 such contracts, the firm had developed a solid reputation among government officials such as Tom Coolidge, director of a government computers and telecommunications station. Coolidge favored contracts with Dragon because it always provides “us with dependable people at a reasonable cost.”

The Resolution: Dragon Systems bid aggressively. It lost money on the initial contract but made a considerable amount of money on the follow-on work, which was largely completed on a time-and-materials (non-bid) basis. It guessed correctly that once it got into the building, it would uncover a lot of other work that needed to be done and that because of its familiarity with the historic building it would be the preferred supplier.

The Lesson: Bidding on contracts requires an evaluation not only of the costs and margins of the specified work, but also a careful analysis of how it changes the probability that you will get future work. That is not easy to do, but it is necessary in order to make reasonable profits in the world of government contracting.

Ronald T. Wilcox is a professor of business administration at the University of Virginia Darden School of Business and the author of “Whatever Happened to Thrift? Why Americans Don’t Save and What to Do About It.”