April 2026
In theory, cost models are meant to forecast the total cost of ownership, assess risk and explore cost and schedule scenarios—giving teams the information they need to justify strategic decisions throughout the project.
In practice, they can fall short. By the time a report lands on a project manager’s desk, the data is already outdated. Instead of guiding future work, these reports simply reflect the past.
On a megaproject, where budgets are massive and error margins thin, that gap is more than inefficient. It’s a liability.
So what does it actually take to build a cost model that works the way it’s supposed to?
The Real Problem with Backward-Looking Models
Traditional cost reporting tends to be constrained by factors that make it even less useful than it should be.
The first is data quality. Garbage in, garbage out—it’s an old principle that still holds true. When source data is incomplete, entered incorrectly or comes from unsynced systems, bad inputs corrupt the model. You may have the best forecasting intentions, but unreliable underlying data means unreliable output, too.
Then there’s fragmentation. Cost data rarely exists in one place. ERP systems, field reports, procurement platforms and schedules each hold part of the picture, but don’t always connect. The result: a model with a partial view.
Overreliance on past data compounds the problem, especially if it’s treated as a blueprint for current or future projects. Past actuals give useful context on historic performance and risks, but don’t reflect current conditions like material price volatility, supply chain issues, labor constraints or site-specific risks, nor do they show the impact of potential risk scenarios.
These process and data problems quietly undermine trust in the cost model’s outputs. When the model consistently misses the mark, trust erodes—if project managers doubt the numbers, they stop using them.
What “Forward-Looking” Actually Means
Forward-looking cost modeling isn’t about having a crystal ball that predicts the future with certainty. It’s about creating a structure around knowing what to anticipate. You do that by building models that incorporate leading indicators—early warning signals that show where delays are looming or where costs are heading—rather than just lagging ones that reflect what’s already happened.
What does this look like in practice? Here are some strategies to layer on top of your existing monthly reporting cycles:
Track committed costs and pending change orders before approval. Imminent, unposted costs are still real. If the model ignores what’s coming, the risk of making decisions on a false sense of solvency rises.
Integrate schedule data to reflect float burn and schedule compression risk in the cost model. A task can only absorb so much delay before it starts affecting the overall timeline. When a built-in buffer is exhausted, the costs of accelerating progress increase. Linking the cost model to the schedule flags financial risks as the schedule tightens.
Shift from monthly reporting cycles to rolling forecasts that update as conditions change. Cadence is the key. Consider how this plays out on a megaproject: that four-week-old snapshot can leave teams behind reality before anyone notices, and after the chance to address it has passed. Rolling forecasts update continuously as new data arrives, so the model stays relevant and actionable.
Think of forecasting as a discipline — and start with better data habits.
How to Build a Forward-Looking Cost Model
There’s a logical sequence to this, and the order matters.
- Audit your data sources first. You can’t fix a model you don’t understand. Identify every system feeding your cost model—ERP, scheduling tools, field reports, procurement—so you know exactly where the gaps and manual handoffs are before you build anything.
- Next, define your leading indicators. With a clear view of your data, identify signals that have predicted cost overruns on similar projects, like procurement delays or rising overtime. This step turns raw data into a watchlist.
- Then separate actuals, commitments and projections. Combining them confuses the picture. Knowing what’s been spent, what’s been obligated, and what’s been estimated is crucial. Treat these as distinct layers for clearer decisions.
- Connect schedule data as a live input. Cost and schedule are inseparable on a megaproject—a shift in one always affects the other. Treating schedule as a dynamic input rather than a separate document keeps the cost model aligned with project reality.
- With the above in place, build a risk and change register linked directly to the cost model. Risks and unresolved change events in separate documents are often ignored. By linking both to the cost model, emerging risks and change impacts surface as cost implications, so there’s still time to act.
- Establish ownership and governance. A solid model needs structure. Designate who maintains inputs, validates assumptions and keeps the model current. Then establish how often the model is reviewed and by whom. Without them, the best-built model can drift.
Overcoming the Cost Model Trust Problem
Building a better model is only half the battle. If project managers don’t trust the outputs, especially after being burned by bad data, they won’t use them. The model loses value.
To gain buy-in, show how numbers are derived, not just what they are. Three practices build transparency:
Version history tracks how the model has evolved over time and, when paired with good documentation, helps users understand why numbers changed rather than just noticing the discrepancy.
Assumption logs document the reasoning behind key inputs: what was assumed, why it was assumed and when it was last revisited.
Clear ownership of inputs means each data point has an accountable person behind it. If something looks off, there’s a person to go to, not just a number to question. That builds accountability instead of ambiguity.
Together, these make the cost model something teams can examine, challenge and ultimately trust.
Cost Modeling as an Ongoing Process
The shift from reporting to forecasting isn’t about buying new software; it’s about rethinking what a cost model is for. Teams that treat cost modeling as a living process rather than a reporting exercise catch problems early enough to address them. That’s the difference between managing a megaproject and reacting to one.
Contruent can help. With its Contruent Enterprise solution, project owners and construction teams can move beyond backward-looking reporting toward cost models that actually forecast—so decisions get made on where a project is going, not just where it’s been. Learn more or request a demo today.