TEV Studies for Agri Projects: What Banks Actually Look For

For capex-heavy agri-projects — food processing, dairy, cold chain, warehousing, farm-equipment manufacturing — the Techno-Economic Viability (TEV) study is the hinge document between the client's operating plan and the bank's sanction. At most public-sector banks, a TEV from an empanelled agency is mandatory for any project finance application above ₹10 Cr; at SBI and the larger banks the threshold is applied rigorously.^[1]

The failure patterns are consistent. Sanctions get delayed or declined not because the project is weak, but because the TEV does not answer the three questions the bank's risk committee will actually ask. This piece explains those questions, what "answering" them looks like, and what separates a TEV that gets sanctioned in two committee cycles from one that gets sent back for revision twice.

What a bank risk committee is actually doing

The risk committee's job is not to evaluate whether the project is a good idea. It is to price the downside. They are answering: what is the probability of non-repayment, and what is the recovery value if non-repayment happens?

Every section of a TEV is either helping the committee answer one of those two questions, or it is decoration. A good TEV is written with the risk committee's job in mind; a bad TEV is written with the promoter's pride in mind.

Question 1 — Is the demand real and sized correctly?

Most agri-projects fail at the demand-assumption layer of the TEV, not the technical layer.

Red flags the risk committee notices instantly:

• Top-down sizing (sector CAGR × region share × time) with no bottom-up tie-out.
• Customer/offtake claims that are not validated by commercial interviews or letters of intent.
• Ramp assumptions that reach design capacity in year 2 of a 5-year ramp curve that the sector does not support.

What strong demand sections look like:

• Bottom-up volume build by product SKU, segment, and customer.
• Named customer evidence — signed LOIs, MOUs, existing dispatch data, or primary interviews at the decision-maker level with at least three anchor buyers.
• A capacity ramp curve that matches what comparable projects in the sector have actually achieved — and names the comparables.

For food-processing projects the anchor buyers are typically modern-trade chains, institutional HoReCa, or export markets; validating offtake at the channel level is what distinguishes a credible forecast from a pitch deck.

Question 2 — Is the capex right-sized?

"Right-sized" means two things: the line is sized for the volumes it will actually ship, not the volumes the promoter hopes for, and the capex does not embed premium-to-market pricing on equipment that depreciates fast.

A rigorous technical assessment:

• Rebuilds capacity bottom-up from line balance, SKU mix, and utilisation assumptions — not from the promoter-supplied capex line items.
• Benchmarks vendor quotes against at least two alternative suppliers per major line.
• Separates tropicalisation and India-specific costs (utilities, dust-control, ambient handling) from the imported-equipment base cost, because under-estimating these is a classic cause of year-one cost overrun.
• Explicitly addresses utilisation — design capacity minus actual throughput minus planned downtime — because a ₹100 Cr line running at 55 percent utilisation is a very different project from the same line at 75 percent.

Right-sizing at TEV stage is one of the most underrated value-adds the consulting agency brings. We routinely find 8-15 percent capex headroom on food-processing greenfields when the line is rebuilt from throughput, which clients either retain as contingency or deploy into higher-return adjacent investments.

Question 3 — What happens in the downside?

A base-case-only TEV is not a TEV. It is a marketing document. Every credible TEV carries at least three scenarios:

Scenario	What it tests	What the committee looks for
Base case	Management plan assumptions	Consistency with historical and market evidence
Upside case	Favourable input costs, faster ramp, better realisation	Not over-engineered — a credible upside band, not a pitch
Downside case	Input commodity stress, ramp delay, large-buyer concentration risk, monsoon impact	DSCR >1.2× sustained through the downside; if not, specific mitigants named

The questions the risk committee will ask on downside:

• What DSCR survives a 10 percent input-cost shock?
• What DSCR survives a six-month delay in the ramp?
• What is the single largest customer concentration, and what happens if they exit?
• Is there a seasonal working-capital peak that breaches the sanctioned working-capital limit?

A TEV that pre-answers these — with numbers, not narrative — is a TEV that sanctions fast.

What the empanelled-agency requirement means

SBI and most of the PSU banks maintain empanelment lists for TEV agencies. SBI's guidelines require, among other things, a minimum ₹100 lac average revenue for the last three years and limit an agency to a maximum of three sectors.^[2] Different banks have different lists; a TEV from an agency not on the lending bank's panel usually means starting over.

Two practical implications:

1. Check the panel before commissioning the TEV. Starting with a non-empanelled agency for the lead bank of the consortium is a classic 3-month rework.
2. Larger consortia need the TEV acceptable to the lead bank and to the other members. Most banks accept each other's empanelled agencies; some do not. Agree up front.

Three project archetype assumption frames

Different agri-project types fail their TEV for different reasons. Three archetypes we see most:

Food processing plant (₹100-250 Cr capex band). The TEV should lead with bottom-up SKU-level demand build, not aggregate throughput tonnage. The single most common failure: the promoter's deck shows design-capacity utilisation at 80 percent by year 2; the sector-comparable ramp data supports 55-65 percent. A strong TEV rebuilds the ramp from scratch using named comparable projects (disclosed in the appendix) and lands utilisation assumptions at a credible band. The working-capital sensitivity section matters disproportionately here — finished-goods inventory plus receivables against modern-trade buyers can peak at 120 days of cost-of-goods at quarter-end, and bank limit-setting needs to account for the peak, not the average.

Cold chain / multi-site warehousing (₹80-200 Cr band). The critical TEV input is occupancy by anchor vs spot customer mix. Promoter decks often assume 75 percent anchor occupancy (higher realisation, lower risk) and 20 percent spot (lower realisation, higher risk) by year 2. Sector data usually supports a slower anchor ramp unless pre-commitment agreements with named anchors are on file. The TEV should explicitly list the anchor-commitment evidence in a dated table — LoIs, signed MoUs, dispatch history — and scenarios should flex the anchor-ramp assumption. PMKSY subsidy eligibility, if claimed, also needs to be verified against the current operational guidelines.^[5]

FPO aggregation / primary processing centre (₹10-30 Cr band, NABARD-refinance eligible). Smaller in capex, bigger in the nuance of the demand side. The TEV needs to validate member-farmer participation assumptions and offtake commitments at the aggregate level, not at the individual-farmer level. The credit-guarantee and equity-grant assumptions under the 10,000 FPO scheme should be included in the capital-stack exhibits.^[6] Because FPO project economics are thin, sensitivity tests on the procurement margin (₹/MT realised by the FPO) matter more than in larger processing projects.

PSU vs private bank TEV expectations

Public-sector and private-sector banks look at TEVs with meaningfully different emphasis, even when the underlying template is similar.

PSU banks (SBI, PNB, Bank of Baroda, Canara, Union, and the regional rural-bank network) are more procedure-driven. A TEV that is structurally incomplete — missing risk matrix, weak working-capital analysis, absent scenarios — will be sent back even if the project is strong. PSU risk committees are often structured around named verticals (agriculture, infrastructure, corporate), and the vertical head's prior experience with similar projects shapes the conversation. Empanelment gating is strictest at PSU banks.^[1]

Private banks (HDFC, ICICI, Axis, and the newer players) are more pragmatic, with a higher tolerance for non-standard project structures and often a faster decision cycle. They rely more on the credit officer's judgment and less on strict template adherence, but they expect sharper quality of analysis on the specific risks the project carries. Private-bank TEVs often include a dedicated section on promoter equity contribution sources and promoter net-worth, which PSU templates treat more lightly.

Syndicated and consortium loans require a TEV that is acceptable to the lead bank and to the other members. In practice, this means writing to the stricter template — which is usually the PSU template if there is a PSU in the consortium — and being prepared for side questions from individual private-bank members on specific risk dimensions.

Common TEV failure modes we see

Sensitivity analysis — where it goes wrong

Sensitivity analysis is the section of a TEV where craft shows up most clearly. Three common patterns we see that cause rejection:

Uniform percentage haircuts. "Revenue -10 percent, EBITDA -20 percent" is not a scenario — it is a marketing convenience. A credible sensitivity test names a specific trigger (a commodity-price spike, a monsoon event, a buyer concentration shock) and lets the model propagate. The DSCR of a pure uniform-haircut case tells a risk committee nothing about project resilience to real-world shocks.

Discount rates that stay constant across scenarios. If the downside scenario represents a materially more stressed project, the cost of capital embedded in the discount rate should usually move. TEVs that hold WACC at a single number across base and downside cases are not properly stressing the IRR.

Missing working-capital sensitivity altogether. For seasonal agri-processors, working capital is often the binding constraint, not EBITDA. A TEV that runs EBITDA sensitivities without a matching working-capital-cycle sensitivity is under-reading the real risk the committee will ultimately fund against.

What a sanction-ready TEV reads like

A TEV that will sanction in two committee cycles:

• Executive summary that leads with the three risk-committee questions and answers them in one page.
• Demand and commercial assessment built bottom-up with named anchor buyers.
• Technical assessment that rebuilds line capacity from SKU mix and utilisation.
• Financial exhibits across base, upside, and two downside scenarios.
• Risk matrix enumerating 5-7 named risks with specific mitigants, monetised where possible.
• Working-capital sensitivity across the seasonal cycle.
• References to comparable projects and their actual ramp curves.

The report is typically 80-120 pages including appendices. Shorter is usually under-supported. Longer is usually padding the risk committee will skim past.

How AgPro helps

We run bank-format TEVs for food-processing, dairy, cold-chain, warehousing, and farm-machinery projects across public-sector banks and leading NBFCs. Our TEV practice is built by ex-banker and ex-industry hands rather than audit-firm generalists, which shows up in how the downside questions get pre-answered. Sanction timelines on our recent work average two committee cycles.

Explore our TEV Studies practice →