₹4.92 lakh crore.
That's the cost overrun sitting across India's central infrastructure pipeline as of June 2026. Everyone talks about that number. Almost no one asks where it actually started.
It didn't start on a construction site. It started in a store register nobody reads, one wrong entry at a time.
The number everyone quotes
The Ministry of Statistics and Programme Implementation (MoSPI) was tracking 1,847 central infrastructure projects worth ₹150 crore and above as of June 2026. Cumulative cost overrun on that pipeline stood at ₹4.92 lakh crore, against a combined original cost of ₹35.61 lakh crore. (Business Standard)
That number gets quoted often. What it doesn't tell you is why. It's a symptom, not a diagnosis.
The number nobody quotes
For a diagnosis, look at the Steel Authority of India Limited (SAIL). The Comptroller and Auditor General (CAG) ran a performance audit on SAIL's inventory management, tabled in Parliament in July 2025. It found SAIL carrying an average inventory of ₹21,698 crore between 2016-17 and 2022-23, about 67 per cent of its current assets, with no fixed benchmark for carrying cost per tonne. (CAG press release)
That's the headline. The detail is where it gets interesting.
Blast furnaces at Rourkela, Bokaro, and Durgapur went off-blast for want of raw material stock, iron ore, coke, sinter, costing 9.32 lakh tonnes of lost Hot Metal production and ₹1,231.52 crore in potential revenue. (Same source)
Purchase orders took longer than the stipulated 186 days from indent to placement in nearly 1 in 10 cases across seven years. Not once. Repeatedly, across the entire audit period. (Same source)
In 2020-21, SAIL lifted 152.73 per cent of its Fuel Supply Agreement quantity from Bharat Coking Coal Limited, despite cheaper coal being available from other suppliers under existing agreements. That excess, 0.17 million tonnes, cost ₹4.65 crore in avoidable expenditure. (Same source)
From 2016 to 2023, SAIL's steel plants consumed imported coal beyond management's own fixed norms. Imported coal costs more than indigenous coal. That single pattern, buying more of the costlier material than the rules allowed, added up to a potential ₹2,539.68 crore in extra expenditure. (Same source)
Read that last one again. Not a shortage. Not a delay. A norm that existed, on paper, and wasn't followed, on the ground, for seven years.
This is not a SAIL problem
No board meeting debates which coal supplier gets used this month. No director signs off on whether imported coal exceeds the norm by two percentage points. These are indent-level decisions. Stores-level decisions. The kind made by someone filling a form, under pressure, on a Tuesday, with no one checking whether the form matches the benchmark.
That's exactly why no audit catches it in time. MoSPI tracks cost and schedule. The CAG's railway logistics audit, currently running across nine Railway Zones and due before Parliament in the 2026 Monsoon Session, will look at process and compliance. (CAG press release) Neither one sits in a store's yard checking whether the coking coal invoiced matches the coking coal received, at the grade and from the origin the contract actually specifies.
That gap is where the money leaks, quietly, for years, until a blast furnace goes cold or an auditor finally opens the ledger.
Where this sits inside a rail EPC contract
Design, Supply, Erection, Testing and Commissioning (DSETC) contracts, the standard model for Indian Railways and Rail Vikas Nigam Limited (RVNL) infrastructure work, carry Supply as a separate pay schedule inside a single lump-sum Engineering, Procurement and Construction (EPC) contract. The contractor buys the full Bill of Quantities (BOQ) early, often against a Supply milestone payment, then holds and stages that material at site stores for the next 18 to 36 months.
Multiply SAIL's pattern, benchmarks on paper, drift on the ground, across 65+ sites instead of one. Each site has its own stores officer, its own local practice, its own private definition of "compliant." The exposure doesn't just add up. It compounds.
Where a Big 4 deck stops, and where the work actually starts
Every finding above answers one question: is there a problem? None of them answer the one that matters: what are you doing about it?
That's usually where the engagement ends. A large firm hands you a deck. Well designed, properly benchmarked, and largely theoretical. It reads well in a boardroom. It rarely survives contact with a stores yard.
I rebuilt stores discipline the other way, across 60-plus live railway infrastructure project sites for a publicly listed rail infrastructure company. I walked the floor. I talked to every stores officer, every site engineer, every procurement lead, because the truth about what's happening in a warehouse lives with the people running it, not in a slide.
That engagement produced a 164-page Stores Management Manual, still the working reference across those sites, and put me in the chair as the domain Single Point of Contact (SPOC) for the Enterprise Resource Planning (ERP) Inventory Module rollout.
Thirty-one years in the Indian Army's Service Corps, followed by years running build-to-suit logistics services and heading a logistics cluster for a listed logistics major, taught me the same lesson twice over. Governance that hasn't been tested on the ground isn't governance. It's paperwork.
Not a template. A way of working.
Zentra Consulting doesn't arrive with a fixed model and stamp it onto every client. No two store operations run the same way, even inside the same company. What doesn't change is the discipline behind the work.
Walk the ground. Walk every site, every warehouse gate, every store's desk, before recommending anything. Talk to everyone who touches the material, not just the people who signed the contract.
War-game it. Stress-test every proposed change against real scenarios before a single line goes into a manual. An audit landing tomorrow. A site shutting down mid-execution. A vendor defaulting on a critical delivery. If the plan doesn't survive the war-game, it doesn't survive the site.
Pilot it live. Prove the approach on one real site, under real conditions. Not in a workshop room. Not on a slide.
Stay past the handshake. Implementation isn't the finish line. I stay through rollout, tuning the framework as compliance requirements shift and the ERP catches up with the ground reality.
Where this leaves you
The audits are done.
The findings are public.
What your organisation does next is a choice, and it's one every Board, Independent Director, CFO, Head of SCM, Head of Procurement, and Head of Projects reading this can make, before an auditor finds the same drift sitting in their own stores register.
If you are executing DSETC or comparable EPC contracts and want to know where your stores discipline actually stands, before someone else tells you, let's talk.