A Bankable Project Report: What Goes In It, and Where They Fail

A project report is the document that converts an intention into something a lender can underwrite. It has a conventional structure, and that structure exists because each section answers a question the credit officer is required to form a view on.

The reports that fail rarely fail on the economics. They fail on internal consistency — a number in one table that cannot be reconciled with the same number elsewhere — and on unsupported assumptions. Both are entirely avoidable.

The structure, and what each part is for

1. Executive summary

One page. The promoter, the activity, the location, the total cost of project, the means of finance, the facility sought, the security offered, and the headline DSCR. An officer reads this to decide how to read the rest.

2. Promoter profile

Who is behind this, and what makes them capable of running it. Qualifications, relevant experience, existing businesses and their performance, net worth with supporting statements, and credit history. Where the promoter already runs a business in the same line, its audited results are the most persuasive page in the document.

3. The project

What is being set up or expanded, where, and why there. Land and building — owned, leased, or to be acquired, with the title position. Plant and machinery, itemised. Utilities: power load sanctioned or applied for, water, effluent treatment. Statutory approvals needed and their status — pollution consent, factory licence, FSSAI, drug licence, fire NOC, whichever apply.

An approval that is "being applied for" with no application number is a risk the officer will price. State the real position.

4. Market and marketing

Demand for what you are making, who buys it, at what price, and how you reach them. Competition, named. Any firm commitments — letters of intent, supply agreements, existing customer relationships — belong here as annexures.

This is the section most likely to be padded with industry-report statistics. A single named customer with a letter of intent is worth more than ten pages of sector growth rates.

5. Technical feasibility

Manufacturing process, capacity, and the basis of the capacity figure. Machine-hours, shifts, yields, wastage. Capacity utilisation assumed in year one and the ramp thereafter.

6. Cost of project and means of finance

Two tables that must sum to the same figure. This is the arithmetic spine of the report.

Cost of project typically comprises:

• land and site development;
• building and civil works;
• plant and machinery, with freight, insurance, duties and erection;
• miscellaneous fixed assets;
• preliminary and pre-operative expenses;
• provision for contingencies;
• margin money for working capital.

Means of finance:

• promoter's capital, with the source identified;
• unsecured loans from promoters or family, and whether they are subordinated;
• term loan sought;
• any subsidy or grant, stated as applied-for rather than received unless sanctioned;
• lease or deferred credit.

Three rules. Every material cost line needs a quotation or a valuation behind it, annexed. Contingency should be a stated percentage, not a plug. And margin money for working capital belongs in the cost of project — leaving it out is why projects run out of cash in month three with the machinery installed and no money to buy raw material.

7. Financial projections

Running to the end of the repayment period:

• projected profit and loss, with the revenue derived from capacity × utilisation × price;
• projected balance sheet;
• projected cash flow;
• depreciation schedule tying to the fixed asset cost;
• term loan repayment schedule, with the moratorium stated;
• working capital assessment.

Each of those has to tie to the others. Depreciation in the P&L equals the depreciation schedule. Interest equals the repayment schedule. Closing cash in the cash flow equals cash in the balance sheet. Reserves movement equals profit after tax less dividends. An officer checks these, and a break costs more credibility than a thin margin would.

8. The ratios

Computed, shown, and commented on:

• DSCR, year by year and average. Average around 1.5 to 2 is the usual expectation, with no single year below roughly 1.25. Show the worst year and explain it;
• debt-equity ratio;
• break-even point, in units and as a percentage of capacity;
• current ratio, TOL/TNW, and return on capital employed;
• a sensitivity analysis — what happens to DSCR if selling price falls 10%, if raw material rises 10%, if utilisation is 10 points lower. Including this unprompted signals that you have stress-tested your own case.

9. Annexures

Quotations, title documents, approvals and applications, promoter net worth statements, audited accounts of existing businesses, letters of intent, CIBIL reports, and the detailed workings behind every projection table.

The failures, ranked by how often we see them

1. Cost of project does not equal means of finance. Simple, fatal, and surprisingly common once a figure has been revised in one table and not the other.
2. Capacity cannot be traced to the machinery. The projections assume an output the quoted machines cannot produce, or produce only on three shifts that the manpower plan does not staff.
3. Year-one utilisation is implausible. Anything above about 60% for a greenfield project needs a specific reason.
4. Working capital margin omitted from the cost of project.
5. Depreciation and interest do not tie to their own schedules.
6. Promoter contribution is not demonstrably available. A number in the table with no bank statement, no asset to be sold, no sanctioned source.
7. The moratorium is ignored in the DSCR computation, so year one looks comfortable because nothing is being repaid in it.
8. Statutory approvals overstated. "Obtained" where the position is "applied for".
9. No sensitivity analysis, leaving the officer to construct the downside case themselves — which they will do less generously than you would.
10. Industry statistics in place of named demand.

How to build one that holds

Build the financial model first, in a spreadsheet, with assumptions on a single sheet that everything else references. Then write the narrative around it. Reports written the other way round — narrative first, numbers assembled to match — are the ones that fail consistency checks, because the numbers were never derived from anything.

Then test it yourself, as the officer will:

• change the selling price by 10% and see whether DSCR survives;
• check that every number appearing in two places appears identically;
• confirm every cost line has an annexure reference;
• read the capacity assumption and ask whether you would believe it from a stranger.

Where Fastlegal fits

We build project reports and the financial models under them on [Fastlegal Cred](https://cred.fastlegal.in), which is our dedicated platform for project reports and CMA data — the model, the ratios, the sensitivity cases and the annexure structure in the format lenders expect. If you already have a report that has come back with queries, we also review and rebuild existing files, which is usually faster than starting again.