Esc

↑↓ move↵ openIndex · Pagefind

GAO and DCMA schedule reviews: use metrics to investigate the plan

Understand the role of schedule-quality metrics, check their populations and thresholds, and connect results to a broader reliability review.

Charley Forey2 min readSchedulingSchedule forensics
On this page
  1. Ask for the exact metric specification
  2. A denominator example
  3. Connect metrics to the broader review
  4. Avoid score-driven repairs

A schedule-health score is a starting point for investigation. It can highlight missing logic, unusual durations or date problems, but it cannot establish that the crews, access and materials will be available. Review the metric definitions and the construction plan together.

The GAO Schedule Assessment Guide presents ten best practices for developing and maintaining a reliable schedule. The familiar DCMA 14-point terminology refers to a separate family of schedule metrics. Do not treat the two labels as interchangeable or assume that either automatically defines your project’s acceptance criteria.

Ask for the exact metric specification

For any tool advertising a DCMA assessment, request the document/version implemented, each formula, its population, exclusions and thresholds. A name on a dashboard is not enough to reproduce a result. The Schedule Forensics source material discusses the historical 2012 DCMA-EA PAM 200.1; historical implementation choices should not be presented as proof of current universal requirements.

Metric question Why it changes the result
Which activities are counted? Completed work, milestones and summary records can change the denominator
Which relationships are counted? External or unresolved links may be treated differently
Which duration is tested? Original and remaining durations answer different questions
Which units are used? Working and calendar days are not interchangeable
Where does the threshold come from? Tool defaults may differ from incorporated project requirements

A denominator example

Imagine a fictional file with 100 activities, of which 60 belong to a defined review population. Six activities in that population have the condition being measured. Dividing by all 100 gives 6%; dividing by the specified 60 gives 10%.

03 / METRIC DEFINITIONKeep the denominator visible.

Same findings. Use the population required by the metric definition, with exclusions disclosed.

Illustrative denominator comparison: six findings divided by all 100 activities equals 6%; the same six findings divided by the defined review population of 60 equals 10%. These percentages are examples, not acceptance thresholds.

Both arithmetic operations are simple. Only one matches that metric definition. A report should expose the six records and the denominator, not just color the result red or green. The percentages here illustrate counting; they are not recommended thresholds.

Connect metrics to the broader review

An open-end count prompts a logic check. A long-duration finding prompts a scope and production discussion. A high-float result prompts investigation of links, calendars, constraints and the milestone being measured. A date warning prompts comparison with actual progress records.

Then ask questions the file may not answer: Is the scope complete? Are the assumptions credible? Does the team maintain the plan? Are risks understood? Can field records support the status update?

Avoid score-driven repairs

Adding arbitrary links or splitting activities without a construction rationale can improve a metric while weakening the model. Document why each revision represents the work more accurately. Recalculate, review milestone effects and retain the before/after evidence.

Use the schedule audit checklist to organize findings, standards crosswalk guidance to establish authority, and the resource library for the complete path.

Share this article

Written by

Charley Forey

Founder of os.construction. We're building the operating system for construction. Project Financials follows job money, and company Accounting covers the books and controls. AI agents remain planned. We write about how money and data actually move on a job, and we build in public.

Apply to the founding program →Start a free trial →
Open file