Capital efficiency is the largest unmeasured lever in capital-intensive industries
Cost programs squeeze a few percent out of each project. Capital efficiency programs change which projects get funded — and the math says the second lever is 5-10x larger than the first. The reason most organizations work the smaller lever is that nobody knows how to quantify the larger one. Marginal NPV per dollar of capex isn't a number on the executive dashboard. The implicit answer is whatever survives the capital committee.
Capital efficiency software fixes that. It computes the marginal NPV of every project, surfaces the projects that are quietly dragging the portfolio average down, and recommends a rebalanced allocation with quantified efficiency uplift. The capital committee gets the same input the CFO has been asking for in different language for years.
What capital efficiency software actually computes
Four metrics, each one designed to expose a specific kind of inefficiency in the current allocation.
Marginal NPV per dollar of capex
For each project, the platform computes the next dollar's worth of NPV. Low marginal NPV projects are flagged regardless of how good their absolute NPV looks — they are spending capital that would generate more return elsewhere.
Resource intensity per unit of NPV
Engineering hours, fabrication slots, and management attention are scarce. The platform divides each project's NPV by its consumption of constrained resources. Projects with low NPV-per-engineering-hour are silently capping the size of the rest of the portfolio.
Correlation cost
Projects highly correlated with existing exposure don't add as much portfolio diversification as their standalone NPV suggests. Capital Project AI computes the correlation cost — the gap between standalone NPV and contribution-to-portfolio NPV — and surfaces the projects that are concentrating risk without paying for it.
Stage-gate efficiency
Stage-gated projects consume capital while waiting for the next decision. The platform tracks the cost of optionality vs. the value created by waiting, and recommends gates where the option to proceed has become uneconomic.
Why Capital Project AI
- Quantifies the right metric. Marginal NPV per dollar — not IRR, not absolute NPV, not payback. The metric the CFO has been trying to ask about.
- Surfaces the inefficiencies most teams miss. Correlation cost and resource intensity are invisible to standard portfolio reviews. The platform makes them obvious.
- Built around the rebalance decision. Recommendations are specific: defer this project, accelerate that one, redirect this engineering capacity, and your portfolio efficiency moves by X bps.
- Built by an ex-Shell capital allocator. Founded by an engineer who allocated capital across $800M of delivered megaprojects.
See your portfolio's true efficiency
Upload your project list — get marginal NPV, resource intensity, and correlation cost for every project, and a quantified rebalancing recommendation in under a minute.
Open the Dashboard →What it looks like in practice
An integrated upstream operator runs a $1.2B annual capex program across 28 funded projects in the U.S., West Africa, and Southeast Asia. The program is on plan. Variance reports are clean. The CFO has been asking for two years why the company's return-on-capital lags the peer group by 250 bps despite operating in similar basins.
Capital Project AI runs the portfolio through capital-efficiency analysis. The findings: six projects together consume 22% of capital but generate 8% of marginal NPV; their absolute NPV is fine but their per-dollar return is lowest in the portfolio. Three are highly correlated with existing exposure — their contribution to portfolio NPV is half their standalone NPV. Two consume disproportionate engineering capacity that's preventing acceleration of three high-marginal-NPV projects. The recommendation: defer two of the six, restructure two as joint ventures (which converts the correlation cost into a fee stream), and accelerate three deferred high-efficiency projects with the freed-up engineering capacity. Net portfolio capital efficiency improves by 280 bps; expected portfolio NPV moves up by $410M over the 5-year horizon.
The same engine drives capital project management software at the execution layer and Monte Carlo project simulation on the underlying probabilistic math. For megaproject-scale decisions, see megaproject risk management.
Frequently asked questions
What is capital efficiency, exactly?
Capital efficiency is the return generated per dollar of capital deployed. In capital projects, it's the portfolio NPV per dollar of capex, adjusted for risk. Capital-intensive industries that improve capital efficiency by even 200-300 bps see massive equity-value uplift; the lever is in better allocation across projects, not in cutting per-project costs.
How is this different from cost reduction?
Cost reduction lowers capex on each project. Capital efficiency picks the right projects to fund in the first place. The math suggests the second lever is usually 5-10x larger than the first, but most organizations spend most of their effort on the first.
What does the platform actually do?
It computes the marginal NPV of every project in the portfolio, identifies projects that are dragging down portfolio efficiency (low marginal return, high resource consumption, high correlation to existing exposure), and recommends a rebalanced portfolio with quantified efficiency uplift.