A schedule quality review tests whether a CPM schedule is logically complete, technically sound and usable for forecasting. It goes beyond checking that the finish date meets the contract.
A schedule can pass a quick look and still fail the first time someone relies on it for a change order, a recovery plan or a delay claim. A structured review finds those weaknesses while they can still be fixed.
What a review checks
- Logic and open ends: every activity should have a predecessor and a successor, except the project start and finish.
- Relationship types, lags and leads: mostly finish-to-start links, with lags and leads used sparingly and explained.
- Constraints: date constraints that override logic, especially mandatory ones.
- Calendars: work weeks, holidays and weather allowances that match the contract and the actual plan.
- Durations and sequencing: whether the durations and the order of work are believable for the scope.
- Critical and near-critical paths: continuous, traceable and made of real work.
- Float behavior: excessive positive float, any negative float, and float trends between updates.
- Status integrity: actual dates, out-of-sequence progress and forecasts sitting in the past.
- Contract compliance: required coding, activity types, submission format and narrative.
Commonly used benchmarks
The Defense Contract Management Agency (DCMA) 14-Point Schedule Assessment is a widely used screening checklist. These are guideline thresholds, not automatic pass or fail rules, and many contracts set their own criteria.
- Missing logic: under 5% of incomplete activities
- Leads: none
- Lags: under 5% of relationships
- Finish-to-start relationships: at least 90%
- Hard constraints: under 5%
- High total float (over 44 working days): under 5%
- Negative float: none
- High duration (over 44 working days): under 5%
- Invalid dates: none
- Missed tasks: under 5%
- Critical path test, Critical Path Length Index and Baseline Execution Index: pass, with the indexes at 0.95 or better
What metrics cannot tell you
A schedule can hit every benchmark and still be wrong. Metrics do not tell you whether the sequence matches how the crew will actually build the work, whether procurement and submittal durations are realistic, or whether the weather calendar reflects the site. A reviewer with construction experience reads the logic as a plan, not as a data table.
What a useful finding looks like
A good finding names the activity, the defect and the effect on the forecast. For example: "Activity A1240, Set roof trusses, has no successor, so its 12 days of float cannot be tested against the dry-in milestone. Linking it to dry-in reduces its float to 2 days and makes it a near-critical activity." That is specific enough for the scheduler to fix and for the owner to verify. The example is illustrative.
When to commission a review
- Before a baseline goes to the owner or agency
- After a major change order or re-sequencing
- When the owner rejects or questions a submission
- Before relying on the schedule for an extension of time claim or formal delay analysis
- When an owner or construction manager wants an independent check on a contractor's schedule
What a useful review delivers
Look for findings ranked by their effect on the completion date, with activity IDs so each one can be checked, a clear recommended correction for each, and a re-run of the checks after the fixes. A list of warnings with no ranking gives you work without direction.
HADVEN provides independent schedule review as part of its CPM scheduling services. For the most common technical faults, see common Primavera P6 schedule errors.