Utilization targets are among the most watched metrics in a consulting firm. They're also among the most poorly forecasted. Most firms track utilization accurately — the time-and-billing system knows exactly what happened last week — but their forward-looking utilization forecasts carry error margins wide enough to make capacity planning genuinely unreliable. The forecast is typically just the pipeline: confirmed engagements through their contractual end dates, plus estimated probability-weighted wins from the opportunity funnel.
What the pipeline forecast doesn't capture is the most consequential variable: whether current engagements will run to their planned end dates, shorten early, or extend. Engagement health signals change the forecasting picture materially — not by adding precision to the pipeline estimate, but by adding early warning about the delivery end of the equation.
Why Standard Utilization Forecasts Are Structurally Late
The standard utilization forecast model treats engagement durations as known quantities. The Hartwell engagement runs twelve weeks, ends on a specific date, the team rolls off on schedule. In practice, engagement durations are variable in both directions — early terminations and scope changes that truncate engagements, and extensions that extend them — and the standard forecast doesn't model either end of this variance with any early-warning capability.
By the time an engagement is clearly heading toward early termination — the client is pulling back, the timeline is compressing, scope is being reduced — the utilization impact is two to four weeks away. For a firm managing eight to twelve active engagements simultaneously, having four of them simultaneously diverge from planned durations (some shortening, some extending) produces utilization swings that can't be managed reactively.
The operations team running the staffing spreadsheet finds out about these divergences when they happen, not before. The engagement health signal layer provides that earlier visibility — typically three to five weeks before the utilization impact materializes — which is enough lead time to begin adjusting staffing plans.
The Engagement Health Signal as a Utilization Predictor
There are three specific engagement health patterns that have reliable predictive relationships with utilization outcomes:
Declining health trajectory in weeks two through four — an engagement that shows consistent health score decline in its first month is approximately three times more likely to end early (pre-scope completion) or require a scope restructuring than an engagement that establishes a stable or improving trajectory. The utilization implication: this team may have capacity opening earlier than the engagement plan suggests. A staffing pipeline that has resources committed through the planned end date may be over-committed if early termination probability is elevated.
Deliverable velocity shortfall exceeding 20% of plan — when an engagement is completing deliverables at 80% or less of the pace assumed in the project plan, the likely outcome is either scope reduction or timeline extension, not acceleration in later phases. Both outcomes have utilization implications. The former creates earlier-than-expected availability; the latter blocks availability for longer than expected. Knowing which direction the engagement is trending is four to six weeks of forecast lead time.
Client engagement signals consistent with disengagement — when the client sponsor's involvement pattern shifts in the way described earlier in this series (reduced touchpoint frequency, delegation of sponsor role to more junior staff), the probability of early engagement termination is elevated. This signal is softer than utilization data but adds confirmatory weight when the other patterns are also present.
How This Changes the Capacity Planning Conversation
A firm managing twelve active engagements, with two showing early warning health signals and one showing strong extension indicators, now has a qualitatively different forecast. The baseline utilization picture — derived from confirmed engagement dates — shows relatively stable forward capacity. The health-adjusted picture shows potential capacity openings in six to eight weeks (from the at-risk engagements) and a capacity constraint in ten to twelve weeks (from the likely extension).
This matters for the business development function as well as the staffing function. If the at-risk engagements end early, the firm needs to be filling that capacity. If the extending engagement holds key senior resources longer than planned, proposals that would have required those resources in that window need to be managed accordingly.
A technology advisory practice at a 140-person consulting firm that serves primarily financial services and insurance clients began using engagement health signals as a capacity planning input in mid-2024. Within two quarterly cycles, their utilization forecast error — measured as the variance between predicted and actual utilization at the end of each quarter — had roughly halved. The improvement wasn't because their pipeline accuracy got better. It was because their delivery-end accuracy improved: they were seeing engagement duration divergences earlier and adjusting their forecast accordingly.
The Nuance: Health Signals Don't Predict Everything
Engagement health signals are correlated with engagement duration outcomes, but they're not deterministic. An engagement can show early warning signals and still recover to a successful conclusion — because a partner intervened, because the client team changed, because the underlying project scope stabilized. The signal is a leading indicator, not a verdict.
This means firms should use health signals to trigger active management conversations, not to automatically reforecast utilization. When an engagement shows declining health in its early weeks, the right response is a check-in with the engagement lead and a review of whether the trajectory has a likely explanation and management path. Only if the signal persists after investigation should it inform a utilization reforecast.
The value is in having the conversation three to five weeks earlier than it would otherwise happen — not in replacing the conversation with an automated update.
A Note on Engagement Extensions
The utilization forecasting benefit flows in both directions. Firms systematically underforecast utilization upside from extensions, not just downside risk from early closures. Engagements that are tracking strongly on health signals in their final month have a materially higher extension probability than average. Surfacing this pattern — "this engagement is in the top quartile of our historical health distribution at this project phase, and engagements in this range extend at approximately twice the base rate" — gives the account partner a useful conversation opener with the client.
The utilization planning implication: capacity that looks like it opens up in six weeks based on the contracted end date may need to be held tentatively for the extension scenario. Firms that begin treating high-health engagements as extension-probable, not just completetion-probable, carry less capacity planning risk over time.