MGMT 8009M Week 10 Debiasing Design Example

Three Repairs Built Into the Bid Gate: A Debiasing Design for a Construction Firm's Committee That Approved Margins Its Projects Never Reached, With an Authentic Dissenter, a Premortem and a Margin Log, Each Owned and Priced

Student Name

Walden University

MGMT 8009M: Organizational Decision Making and Judgment

Instructor Name

Month Day, Year

Three Repairs Built Into the Bid Gate: A Debiasing Design for a Construction Firm's Committee That Approved Margins Its Projects Never Reached, With an Authentic Dissenter, a Premortem and a Margin Log, Each Owned and Priced

Four Losses in Six

Sutter Point Constructors, a composite engineering and construction firm working on water treatment and industrial process plants, lost money on four of its last six fixed-price bids. In each of the six, the bid committee approved a margin the finished project did not reach. The pattern is consistent enough to describe in a sentence: approved margins ran above realized margins on every project, and the gap was widest on the most competitive bids.

[after]

Four of the six realized margins fell below the 3 percent level the firm treats as the cost of overhead, which is why the firm counts them as losses. The two bids with the fewest competitors lost the least margin.

Table 1

Approved and Realized Margins on the Last Six Fixed-Price Bids

Repairs Built Into Routine

Heath et al. (1998) argued that organizations can compensate for individual judgment errors through cognitive repairs: practices, procedures and norms that correct for predictable biases without requiring individuals to become less biased. Their argument shifts the target of debiasing from minds to routines. That shift matters for Sutter Point, because the firm had already scheduled a two-day training course on cognitive bias for its estimators and committee members. Training addresses individuals and assumes that awareness will change judgment under deadline and competitive pressure. The design below instead builds three repairs into the bid committee's final gate, where the margin is approved.

The Design

[after]

Every row traces to the record. The first two address the approval meeting, where optimism concentrated on competitive bids. The third addresses the reason the pattern persisted for six bids: the committee never saw its own accuracy.

Table 2

Three Countermeasures by Target Failure

A Dissenter Who Meant It

The obvious repair is a devil's advocate, a member assigned to argue against each bid. Nemeth et al. (2001) compared assigned devil's advocacy with authentic dissent, in which the dissenter actually holds the opposing view. Authentic dissent stimulated more original thinking and more consideration of alternatives in the majority than role-played dissent did, and role-played dissent could even lead the majority to strengthen its initial position. A committee that knows the critic is playing a part can discount the critique as theater.

The design therefore does not assign a rotating critic. For each bid above $10 million, the chief estimator identifies an estimator who argued, at any earlier stage, for a lower price, a higher contingency or declining the bid. That person presents the case for declining or repricing at the final gate. If no one on the estimating team held such a view, the record of that fact goes into the bid file, which is itself informative.

A Premortem at the Last Gate

Before the vote, each committee member writes, in three minutes and without discussion, the most likely reason the project will lose money if it is won at this margin. The chair collects the answers and reads them aloud before debate opens. The technique draws on research by Mitchell et al. (1989), who found that people asked to imagine that an event had already occurred generated more reasons for it than people asked to imagine that it might occur. Framing failure as a fact makes its causes easier to find. The design fixes the premortem as the first agenda item at every final gate, so that it cannot be skipped on a busy week.

A Log That Can Be Checked

Each approved bid records its predicted margin and the three reasons the committee gave for it. At closeout, project controls compares the realized margin with the prediction and notes which reasons held. After two years, the log will give the committee its first record of its own accuracy, by bid type and number of competitors. The log does not change any single decision. It changes what the committee knows about itself.

Why Training Alone Falls Short

The scheduled training is not harmful, and the design does not cancel it. But knowing about optimism bias rarely prevents it under the conditions that produce it: a competitive bid, a deadline and a team that has worked hard on an estimate. The three repairs work regardless of whether members remember the training.

Costs and What Success Would Look Like

Together the repairs cost about four hours of staff time per bid and two hours per project at closeout. Sutter Point submits about fifteen fixed-price bids a year. Success would mean a narrowing gap between approved and realized margins on competitive bids, visible in the log after two years, and possibly fewer bids won at thin margins.

Limits

Sutter Point is a composite, and six bids are too few to estimate the size of the bias with confidence. The log is designed to supply the evidence the firm currently lacks.

References

Heath, C., Larrick, R. P., & Klayman, J. (1998). Cognitive repairs: How organizational practices can compensate for individual shortcomings. Research in Organizational Behavior, 20, 1-37.

Mitchell, D. J., Russo, J. E., & Pennington, N. (1989). Back to the future: Temporal perspective in the explanation of events. Journal of Behavioral Decision Making, 2(1), 25-38. https://doi.org/10.1002/bdm.3960020103

Nemeth, C., Brown, K., & Rogers, J. (2001). Devil's advocate versus authentic dissent: Stimulating quantity and quality. European Journal of Social Psychology, 31(6), 707-720. https://doi.org/10.1002/ejsp.58