Level-of-effort work is the scheduling and EVMS category that project controls professionals talk about the least and that auditors examine the most. Used correctly, LOE is a legitimate and necessary planning construct. Used incorrectly — and it is frequently used incorrectly — it is one of the most effective tools available for masking schedule and cost performance problems without technically violating any individual rule.
What LOE Is Supposed to Be
Level-of-effort is a type of work whose progress cannot be measured in terms of discrete physical output. It is effort that supports the program — project management, system engineering oversight, safety monitoring, quality assurance surveillance — without producing a deliverable that can be objectively measured at a point in time. The earned value for LOE work is tied to the passage of time: as the period covered by the LOE budget passes, the budget is automatically earned. LOE work earns at the rate it is planned, regardless of how much or how little was actually accomplished.
This is the defining characteristic of LOE that makes it both necessary and dangerous. It earns automatically. There is no independent test of whether the work was actually performed or whether it was performed effectively. Budget flows through LOE accounts on schedule by definition.
EIA-748 guidelines (Guideline E10) require that LOE be clearly distinguished from discrete work and apportioned effort. DCMA surveillance practice treats LOE concentrations above approximately 15% of total control account effort as a surveillance concern, because above that threshold the automatic earning characteristic of LOE can meaningfully distort overall program performance metrics.
How LOE Gets Misclassified
The misclassification of discrete work as LOE is one of the most common and most consequential EVMS violations. It takes several forms.
Burying troubled work packages in LOE. When a discrete work package is behind schedule and the CAM anticipates negative variance, reclassifying the work as LOE eliminates the problem — not by solving it, but by making it disappear. LOE work always earns on schedule. A discrete task that is two months late produces a significant schedule variance. The same task reclassified as LOE shows no variance at all.
Using LOE as a budget reserve. LOE accounts that are larger than the supporting effort actually requires provide a reservoir of automatically earned value that can offset negative variance in discrete work areas. The program's overall SPI and CPI look better than they are because a significant portion of the earned value is arriving on autopilot.
Misclassifying material or fabrication as LOE. Physical work that produces measurable output — fabrication, installation, testing — is discrete work by definition. Treating it as LOE because measurement is inconvenient, or because the work is difficult to status, is a misclassification that DCMA surveyors specifically look for.
Planning package conversion avoidance. Work that should have been converted from planning packages to discrete work packages, but hasn't been, may be reclassified as LOE as a workaround — avoiding the discipline of detailed work package planning while continuing to earn budget against a vague scope definition.
The 15 Percent Threshold
DCMA surveillance practice uses 15% of total control account effort as a rough threshold for heightened LOE scrutiny. Above this level, the proportion of automatically earned value is large enough to meaningfully mask discrete work performance. A control account where 40% of the budget is in LOE can show acceptable overall performance even if the discrete work portion is significantly behind schedule and over cost — the LOE earnings are smoothing the reported metrics.
The threshold is not a hard rule, and the appropriate LOE percentage varies by control account type. A project management control account legitimately carries high LOE. A construction activities control account that shows 40% LOE has likely been misclassified.
What to Look For in the Schedule Data
LOE misclassification is detectable in P6 and MS Project environments when schedule data is examined carefully.
Activities marked as LOE that have defined physical outputs — drawings, specifications, equipment deliveries, concrete pours — are candidates for misclassification review. LOE activities with durations that perfectly match the reporting period, suggesting they were created to absorb budget rather than to represent real work, deserve scrutiny. Control accounts where the LOE percentage increased significantly between baseline and the current schedule, without a formal scope change, may indicate reclassification of troubled discrete work.
The most revealing analysis compares LOE percentage against actual performance trends in the corresponding discrete work. A control account whose discrete work CPI has been declining while LOE earnings have been stable is showing exactly the pattern that misclassification produces: real performance problems absorbed by automatic earnings.
The Program Impact
LOE abuse does not just distort metrics. It erodes the management information value of the entire EVMS. When program-level SPI and CPI are partially constructed from automatically earned value that reflects the passage of time rather than the accomplishment of work, those metrics cannot be trusted as early warning indicators. The whole point of earned value — providing earlier and more accurate visibility into program performance — is defeated.
A program manager who adjusts resource allocation, risk posture, or management attention based on metrics that have been smoothed by LOE misclassification is operating on false information. By the time the underlying discrete work problems are large enough to overwhelm the LOE buffer, the window for effective corrective action has often closed.
The Forensic Intelligence Engine flags LOE concentration anomalies, detects reclassification of discrete work, and cross-validates LOE usage against the physical work types defined in the program's contract and Work Breakdown Structure.