# Scenario and Sensitivity Model Development: Cases, Drivers, Output Tables

Published: 30 Oct 2026
Author: Treelife
Practice area: Finance
Tags: DCF sensitivity analysis WACC terminal growth, downside case before management action, output tables for financial model, scenario analysis in financial model, scenario and sensitivity model development, scenario selector in Excel financial model, sensitivity analysis in financial modelling, tornado chart sensitivity analysis
Source: https://treelife.in/finance/scenario-and-sensitivity-model-development/

---

Blog Content Overview

- [0.1 What is the difference between scenario analysis and sensitivity analysis?](#What_is_the_difference_between_scenario_analysis_and_sensitivity_analysis)

- [1 What model architecture does scenario and sensitivity model development need?](#What_model_architecture_does_scenario_and_sensitivity_model_development_need)
- [2 How do you define cases for financial model scenario analysis?](#How_do_you_define_cases_for_financial_model_scenario_analysis)
- [3 Which drivers belong in the scenario layer?](#Which_drivers_belong_in_the_scenario_layer)

[3.0.1 Need a scenario layer built into your financial model? 
Let’s Talk](#Need_a_scenario_layer_built_into_your_financial_model_Let8217s_Talk)

- [4 How do you build a scenario switch in Excel?](#How_do_you_build_a_scenario_switch_in_Excel)
- [5 How do you run sensitivity analysis in financial modelling so readers trust it?](#How_do_you_run_sensitivity_analysis_in_financial_modelling_so_readers_trust_it)
- [6 Which output tables should the model produce?](#Which_output_tables_should_the_model_produce)
- [7 How do scenario outputs feed valuation under Indian rules?](#How_do_scenario_outputs_feed_valuation_under_Indian_rules)
- [8 Common mistakes that cost founders time and money](#Common_mistakes_that_cost_founders_time_and_money)
- [9 How Treelife approaches scenario and sensitivity model development](#How_Treelife_approaches_scenario_and_sensitivity_model_development)
[9.1 How do you start scenario and sensitivity model development in the next 30 days?](#How_do_you_start_scenario_and_sensitivity_model_development_in_the_next_30_days)

- [10 FAQs on Scenario and Sensitivity Model Development](#FAQs_on_Scenario_and_Sensitivity_Model_Development)

[10.0.0.1 Regulatory references](#Regulatory_references)

Scenario and sensitivity model development is the work that turns a single-line forecast into a range an investor, lender or board can price. In India the work carries extra weight because funding rounds, venture debt and cross-border share pricing all ask the same question: what happens when the plan is wrong? A good build separates the cases from the drivers, keeps one model behind every output, and presents results in tables each audience can read in a minute. This article sets out how to do that, step by step, with worked numbers.

### What is the difference between scenario analysis and sensitivity analysis?

Sensitivity analysis changes one input, or two, and holds everything else constant to show which assumption moves an output most. Scenario analysis changes several inputs together to describe one coherent future, such as a slow-demand year with longer collection cycles. Sensitivity diagnoses the model for the builder. Scenarios communicate a decision to the reader. A model needs both, and each needs its own mechanics.

## What model architecture does scenario and sensitivity model development need?

Use one workbook with five layers: a control sheet, an input sheet holding every case side by side, a calculation engine that reads only the live case, a checks sheet and an output sheet. The scenario switch sits on the control sheet and nothing in the engine is typed over. One model with many cases is what lets a reviewer trust every number.

The precondition is a [driver-based financial model](https://treelife.in/services/virtual-cfo/): revenue, cost and cash lines calculated from operating drivers such as customers, price and debtor days, not typed-in growth percentages. Without drivers there is nothing for a case to change. The three-statement model should already link income statement, balance sheet and cash flow before any case is layered on top.

The five layers each have one job:

- **Control:** the scenario selector, model date, forecast start date, unit switch (₹ lakhs or ₹ crore), case probability weights and a master check flag.

- **Inputs:** every driver in its own row, one column per case, plus a Live column that returns whichever case the selector points to and an optional Override column for analyst tests, left blank in any copy that goes to a board.

- **Engine:** the monthly or quarterly calculation of revenue, cost, working capital, capex, tax, debt and cash. It reads the Live column and nothing else.

- **Checks:** balance sheet balances, cash ties to the cash flow statement, the selector sits inside its valid range, and no engine cell holds a typed number.

- **Outputs:** the case comparison, sensitivity grids, valuation grid and lender tables.

The forecast start date is the control most models miss. Actual months before it are locked and identical in every case, and cases diverge only after it. Without it, a downside run in October 2026 quietly restates April to September. Load closed actuals after each month-end and move the date forward one month.

Separate workbooks for each case look simpler and fail slowly.

Two Indian conventions belong in the architecture on day one:

- **Calendar:** align the model to the financial year from April to March, and keep a quarter-end view because boards and lenders commonly review quarterly.

- **Currency:** carry an INR/USD row as a driver for any export revenue or imported cost instead of burying a rate inside a formula.

**Scenario and sensitivity compared by build and use**

DimensionSensitivityScenarioInputs changedOne or two, others held constantSeveral together, moved as a setQuestion answeredWhich assumption moves the output most?What does this future look like end to end?Built withData tables, tornado ranking, breakeven solveCase columns, selector, narrative per casePrimary readerThe model builder and the reviewing analystThe board, investor or lender making a decisionOutput formatGrids and ranked swing tablesSide by side case comparisonUse whenRanking drivers, validating the model, deciding where to buy better dataPlanning, lender and board decisions, stress testsCannot tell youHow drivers interactWhich single assumption matters most, and it carries judgement biasCommon failureEqual percentage ranges that are not equally plausibleCases that change one driver and carry three labels

## How do you define cases for financial model scenario analysis?

Start with base, upside and downside, then split the downside into before and after management action. Add a stress case only for a defined shock, such as a delayed round, or a lender’s bank case when one is required. Each case is a written narrative first and a set of driver values second. Cases that change drivers independently are sensitivities with scenario labels, and a sharp analyst will say so. Getting this right is the centre of scenario and sensitivity model development.

The narrative comes first because it forces the drivers to move together. A downside written as “two large customers delay renewals by two quarters and collections stretch to 80 days” tells you that growth falls, debtor days rise and the cash cycle lengthens in the same period.

Split the downside in two:

- **Before management action:** what happens if the company does nothing. Lenders ask for this version.

- **After management action:** the plan the board would actually follow, such as a hiring freeze and a collection drive from the first quarter. Boards approve this version.

Showing only the second hides the cash exhaustion date that the first reveals. The monthly cycle that keeps these cases current is covered in [rolling budget forecast implementation](https://treelife.in/finance/rolling-budget-forecast-implementation/).

Rules that keep cases honest:

- Move correlated drivers together: slower growth usually arrives with longer debtor days and weaker pricing, and a weaker rupee with higher imported cost.

- Do not set every driver at its extreme together. A downside with each driver at its worst is a stress case, and it overstates the risk a board should plan for.

- Name cases by their narrative and sensitivities by the driver moved, such as “Downside before action” and “Revenue growth minus 20%”, so a label tells the reader what was run.

- Keep the stress case for a defined shock, such as the next round slipping by two quarters, and not as a vaguer worse downside.

**Illustrative case definitions for a B2B software company with opening FY2026-27 revenue of ₹40 crore**

DriverDownside before actionDownside after actionBaseUpsideRevenue growth, FY2027-28 to FY2031-3220%, 16%, 14%, 12%, 10%20%, 16%, 14%, 12%, 10%35%, 30%, 26%, 22%, 18%45%, 38%, 32%, 27%, 22%Gross margin58% flat58% flat62%, rising 1 point a year64%, rising 1.5 points a yearOperating cost growth a year16%8%18%22%Debtor days80756055NarrativeDemand slows, collections stretch, no management responseSame market, hiring freeze and collection drive from quarter onePlan as approved by the boardFaster enterprise wins, hiring ahead of revenue

The figures in this article come from a simplified illustrative model with opening operating costs of ₹25 crore, opening cash of ₹12 crore, capex at 3% of revenue and tax at 25% of positive EBITDA. They show mechanics, not benchmarks for any sector.

## Which drivers belong in the scenario layer?

Put a driver in the scenario layer only if it is uncertain, material and observable. Five to eight drivers usually clear that bar. Everything else stays a fixed input on the assumptions sheet. A short driver list is what makes a case readable, auditable and quick to refresh each month.

Apply three tests to each candidate driver:

- **Uncertain:** a plausible range exists that a reasonable person would defend.

- **Material:** a move across that range shifts the output you care about by a meaningful amount, for example more than 5% of enterprise value or more than a month of runway.

- **Observable:** you will know within a quarter whether the driver is moving, so the case can be re-pointed quickly.

List twenty to thirty candidate drivers, run a one-way sensitivity on each, rank them by swing and keep the top five to eight. Choose the lowest-level driver available: if the model carries both revenue growth and new customers multiplied by price, pick one, because two drivers for the same line double count.

**Typical scenario drivers by business model**

Business modelPrimary driversSecondary driversOutput most exposedB2B softwareNew customers, net revenue retention, price, gross marginHiring pace, debtor daysRunway, enterprise valueConsumer and D2CRepeat rate, acquisition cost, return rate, contribution marginInventory days, payment gateway settlementCash, contribution marginManufacturingVolume, realisation, raw material cost, capacity usageInventory days, debtor days, power costEBITDA, working capitalLending and NBFCDisbursement, yield, cost of funds, credit costOperating cost, collection efficiencyNet interest margin, capital adequacyFund managerFundraise pace, deployment pace, fee yieldOperating cost, carry timingManagement fee income, break-even AUM

These are starting lists for scenario and sensitivity model development, not answers. Driver rankings differ by company and by output, so run the ranking on your own model before fixing the list.

Three drivers deserve a place in Indian models more often than they get one:

- **Capital arrival date:** when the next round or debt facility closes and for how much. Runway cases depend on it more than on most operating drivers, and a round that slips by a quarter is a common shock.

- **INR/USD rate:** for companies with export revenue or imported cost.

- **Customer payment behaviour:** large Indian enterprises and government buyers can pay well beyond contract terms, and debtor days is where that shows up.

Add one non-operating input: a management action switch that turns on the hiring freeze, cost cuts and collection drive, so a reviewer sees what the actions are worth.

> 

#### 
Need a scenario layer built into your financial model? 
[Let’s Talk](javascript:void(0))

[

](https://calendly.com/consulttreelife/20min?utm_source=blogbannertreelife)

## How do you build a scenario switch in Excel?

Create a named cell holding the case number, list each driver in rows with one column per case, and add a Live column that returns the chosen case with INDEX or CHOOSE. Point every engine formula at the Live column. Avoid OFFSET and INDIRECT, which are volatile, and keep the engine free of typed numbers.

```
Scenario_ID (control sheet, cell C4) = 2
Live column (inputs sheet, cell H12) = INDEX(D12:G12, 1, Scenario_ID)
Alternative = CHOOSE(Scenario_ID, D12, E12, F12, G12)
Case label (every output header) = INDEX(case_names, Scenario_ID)

```

The build takes five steps:

- Name the selector cell Scenario_ID and restrict it with data validation to whole numbers from 1 to 4.

- Lay out each driver once, one row each, with one column per case and a Live column on the right.

- Point every engine formula at the Live column and never at a case column.

- Print the case label in the header of every output table so no PDF is ever ambiguous.

- Add checks: the selector sits in range, engine cells contain formulas, and case probability weights add to 100%.

Excel’s built-in Scenario Manager should be avoided for anything an outsider will read. It caps at 32 changing cells and overwrites the changing cells, and its summary report pastes static values, so the report goes stale the moment any input changes.

For a side by side comparison that stays live, key a data table on the selector. Place the selector and the comparison table on the same sheet, because Excel requires a data table’s input cell to sit on the same sheet as the table. List 1 to 4 down the left column, link the output formulas across the top row, and set the column input cell to Scenario_ID. Excel recalculates the whole model once per case and fills the grid.

Five controls stop the comparison table from misleading anyone:

- **Tie-out check:** the data table row for the current selector must equal the live engine output. A non-zero difference means the table is stale.

- **Calculation mode:** large monthly models are often set to Automatic except for data tables. If yours is, press F9 before every export and let the tie-out check confirm it.

- **Circularity:** avoid circular references such as interest on average cash. Data tables built over circular models can return unstable or wrong values without an error. Compute interest on opening balances.

- **Version stamp:** put the model date and a version number on the control sheet and in every output header.

- **Debt schedules:** keep contractual repayment schedules, such as [venture debt](https://treelife.in/finance/venture-debt-vs-equity-funding/) tranches, fixed across cases. If a downside re-sculpts repayments, the model shows debt amortising on terms the lender never agreed.

Keep data tables on one output tab so they do not slow the engine sheets. Because Excel needs each input cell on the same sheet as its table, add a flex cell on that tab, set to zero, that is added to the driver in the Live column, and point the table at it. Whichever cell you nominate must be one the model actually reads, or the table returns a flat column of identical values.

## How do you run sensitivity analysis in financial modelling so readers trust it?

Run one-way sensitivities on every candidate driver to rank them, then run two-way grids only on the pairs a reader will question. Use plausible ranges, not equal percentages, show the base cell in the centre, and add a breakeven for the output that matters most. A sensitivity earns trust when its ranges are defensible.

A one-way run produces the tornado ranking, a table or chart that orders drivers by the swing they cause. The ranges should reflect how far each driver can plausibly move. Revenue growth might run from 0.8 times to 1.2 times plan, gross margin might move two points either way, and debtor days fifteen either way. Applying one percentage to every driver looks neutral and is not.

**Tornado ranking of enterprise value in the illustrative base case (base value ₹97.1 crore)**

Driver and rangeLow-case value (₹ crore)High-case value (₹ crore)Swing (₹ crore)Revenue growth, 0.8 times to 1.2 times plan20.5185.8165.3Operating cost growth, plus 3 to minus 3 points54.0136.482.4Discount rate, plus 2 to minus 2 points76.0130.654.6Gross margin, minus 2 to plus 2 points81.2113.131.9Terminal growth, minus 1 to plus 1 point88.5107.919.4Debtor days, plus 15 to minus 1588.4105.817.4

A 20% change in growth rates moves enterprise value by roughly 80% to 90%, because the operating cost base does not flex with revenue. That operating gearing is why growth tops the ranking.

The ranking also depends on the output. Rank the same drivers by lowest closing cash in the downside after action and the order changes. Operating cost growth of 3 points more takes lowest cash from ₹7.0 crore to ₹1.1 crore. Debtor days of 15 more takes it to ₹4.7 crore, and gross margin of 2 points less takes it to ₹5.0 crore. An investor reads the valuation tornado. A lender reads the cash tornado. Build both.

Use the ranking in two ways:

- **Test the model:** an output that does not move when a driver moves points to a hardcode or a broken link.

- **Direct diligence effort:** the driver with the widest swing gets a cohort analysis, and debtor days gets a monthly ageing review.

Two-way grids answer the questions a tornado cannot, because they show interaction. Useful pairs include:

- Revenue growth against gross margin.

- Hiring pace against the capital arrival date.

- Customer acquisition cost against churn, for software and consumer models.

- Discount rate against terminal growth.

Formatting rules that help the reader:

- Place the base case in the centre cell and mark it with a border.

- Use equal steps on each axis and label both axes with the variable name and unit.

- Use conditional formatting to show a threshold, such as cash below a covenant floor, and not as decoration.

- Type the axis values or build them from a separate copy of the base inputs. Never link an axis to the input cell the table is flexing, because Excel then returns wrong results without warning.

Finish with a breakeven, the most useful sentence the model gives a board. Use Goal Seek, or a reverse stress test, to find what has to be true for the output to fail. In the illustrative base case, closing cash turns negative only if gross margin runs about 12 points below plan in every year. The same logic applies to [runway and burn analysis](https://treelife.in/finance/burn-rate-runway-calculation-for-startups-in-india/), where the question is how far a driver can move before cash runs out.

## Which output tables should the model produce?

Produce six output tables: a case comparison, a cash and runway table, a valuation grid, a returns and dilution table, a lender covenant table and a driver bridge. Each answers one audience’s first question, carries the case label, and ties back to the engine through a check cell.

In scenario and sensitivity model development, design the outputs before the engine, working backwards from what each reader will ask. A founder asks how long cash lasts. An investor asks what the business is worth across cases and what their stake becomes. A lender asks how far the covenant is from breach. An operating head asks which driver caused the gap between base and downside.

**Output tables by audience**

AudienceOutput tableMetrics shownCase viewFounder and boardCase comparisonRevenue, EBITDA margin, lowest cash, cash exhaustion yearAll cases side by sideFounder and finance headCash and runwayMonthly closing cash, months of runway, funding gapDownside before and after actionInvestorValuation grid and returnsEnterprise value, probability-weighted value, ownership after the roundEach case, plus weightsLenderCovenant headroomDebt service cover, net debt to EBITDA, headroom to the covenant level in the term sheetEach case, by test dateOperating headsDriver bridgeMovement from base to downside by driver, in ₹ croreBase to chosen caseReviewer or valuerAssumptions registerEach driver, value by case, source, owner, last review dateAll cases

The case comparison is the table most readers will remember. Keep its layout identical for every case, show units in the header, and include a row that names the first year cash turns negative.

**Case comparison output for the illustrative company (₹ crore unless stated)**

MetricDownside before actionDownside after actionBaseUpsideRevenue, FY2027-2848.048.054.058.0Revenue, FY2031-3278.278.2127.3163.7EBITDA margin, FY2031-32-9.1%11.0%21.1%28.7%Lowest closing cash, FY2027-28 to FY2031-32-24.07.011.113.1Headroom of lowest cash over a ₹5 crore minimum buffer-29.02.06.18.1First year closing cash is negativeFY2029-30NoneNoneNoneEnterprise value at 14% discount rate and 5% terminal growthNot meaningful14.697.1195.1

Read the table the way a lender would. Without action the company runs out of cash in FY2029-30. With a hiring freeze and a collection drive from the first quarter it never falls below ₹7.0 crore. The gap between those two columns is the value of management’s response, and it is the number a board should debate. Against the ₹5 crore buffer, the downside after action clears by only ₹2.0 crore, and that thin cushion is the finding.

Four habits keep output tables credible:

- Show a delta to base column when a reader will compare cases.

- Never publish a table without its case label.

- Keep the assumptions register current, because it is the first thing a reviewer opens.

- Link every figure to the engine so a changed input never leaves a stale number in a deck.

For the investor-facing version of these tables, see how investor pitch financial projections are presented.

## How do scenario outputs feed valuation under Indian rules?

Scenario outputs support commercial valuation and FEMA pricing, but not tax fair market value for unquoted equity from 01/04/2026. Rule 57 of the Income-tax Rules, 2026 prescribes a net asset formula and no DCF option. FEMA Rule 21 accepts any internationally accepted pricing methodology on an arm’s length basis, so a documented DCF remains a common basis.

The Income-tax Act, 2025 came into force on 01/04/2026, and the Income-tax Rules, 2026 were notified on 20/03/2026 with effect from the same date. Rule 57 now carries the fair market value rule that Rule 11UA held under the 1962 Rules. For unquoted equity shares it prescribes a formula of the form (A + B + C + D – L) x PV / PE, built from book values and prescribed fair values on the balance sheet. The DCF option in Rule 11UA(2) is not carried across. It existed mainly for Section 56(2)(viib) of the Income-tax Act, 1961, the [angel tax](https://treelife.in/compliance/understanding-valuation-rules-for-share-transfers-post-angel-tax-removal/) provision, which the Finance (No. 2) Act, 2024 omitted with effect from 01/04/2025 for all investors. The Income-tax Act, 2025 does not re-enact it.

Two consequences follow for model builders:

- A projection model no longer drives tax fair market value for unquoted equity, so label DCF outputs as commercial value and keep them out of any tax computation.

- Because the Rule 57 formula reads the balance sheet, the quality of the projected balance sheet matters for planned share transfers. That is one more reason to keep the three-statement structure intact under every case.

**Valuation purpose, governing provision and what the model must do**

PurposeGoverning provisionMethod positionWhat the model must doTax fair market value of unquoted equityRule 57, Income-tax Rules, 2026 (Income-tax Act, 2025, effective 01/04/2026)Prescribed formula on balance sheet values, no DCF optionKeep DCF outputs out of tax computations; keep the projected balance sheet cleanPremium on shares issued to resident or non-resident investorsSection 56(2)(viib), Income-tax Act, 1961, omitted by the Finance (No. 2) Act, 2024 from 01/04/2025; no equivalent in the Income-tax Act, 2025Provision no longer applies to issues from that date; earlier issues stay open to assessmentDefend case weights to investors, not to a tax officerIssue or transfer of unquoted equity with a non-residentRule 21, FEMA (Non-debt Instruments) Rules, 2019Any internationally accepted pricing methodology on an arm’s length basis, certified by a chartered accountant, SEBI-registered merchant banker or practising cost accountantRecord case selection, weights and discount rate so a certifier can follow themNegotiated pricing of a round or secondaryAgreement between the parties, no statutory formulaScenario-weighted DCF, multiples or milestone valueShow the range, the weights and the resulting ownership by case

For commercial valuation, the DCF sensitivity grid is the output readers study first. The table below flexes the discount rate, which is the weighted average cost of capital (WACC), against terminal growth for the illustrative base case.

**DCF sensitivity grid for the illustrative base case, enterprise value in ₹ crore**

WACCTerminal growth 4.0%4.5%5.0%5.5%6.0%12%11612313114015013%10110611211812614%88939710210815%798286899416%7173767983

The grid shows a range of ₹71 crore to ₹150 crore from two inputs alone. In the illustrative base case, terminal value is about 82% of enterprise value, so most of the answer rests on assumptions beyond year five. Show that share next to the grid and cross-check it against a multiple. Weighting the cases at 25% downside after action, 50% base and 25% upside gives ₹101.0 crore against ₹97.1 crore for the base alone. The weights are judgement, so keep them as visible inputs on the control sheet. Place the grid beside the valuation summary, not in an appendix.

Background on how valuations are negotiated is in the guide to [how startup valuation works in India](https://treelife.in/startups/how-startup-valuation-works-in-india/), and the rules for share transfers after the angel tax change are covered in valuation rules for share transfers post angel tax removal.

## Common mistakes that cost founders time and money

Most scenario and sensitivity model development efforts fail in the same five places. Each is cheap to fix before a data room opens and expensive to explain once an analyst has found it.

- **Calling a sensitivity a scenario set.** A model with base, upside and downside columns where only price changes is one sensitivity with three labels. It happens when cases are built from a list of price points. The fix is to write the narrative first and move every correlated driver with it.

- **Overwriting the base case to run a downside.** Nobody can say what the base was last week. It happens when cases are tested by typing over inputs. The fix is one column per case and a Live column driven by the selector.

- **Showing only the downside after management action.** The board sees a comfortable minimum cash balance and never sees the exhaustion date. It happens when mitigation is built into the downside from the start. The fix is two downside columns and a management action switch.

- **Leaving data tables stale or mis-wired.** Tables go stale when calculation skips them, and wrong when an axis links to the input it flexes or the model is circular. The fix is a tie-out check on every table, axes typed or built from a base copy, and interest on opening balances.

- **Presenting a DCF without its terminal value share.** A valuation where terminal value is above 80% of the total invites the first question in every diligence call. The fix is to show the share, add a multiple cross-check and show the grid, not a point.

A model that cannot name its case, tie its tables or explain its ranges turns diligence into a defence of the spreadsheet, as covered in financial model validation before a data room.

## How Treelife approaches scenario and sensitivity model development

Scenario and sensitivity model development at Treelife starts with the reader of the outputs, not the structure of the workbook. Cases, drivers and output tables are agreed in a scoping session, then built on a single control-sheet architecture and tied out with checks before anything is shared externally.

### How do you start scenario and sensitivity model development in the next 30 days?

Start scenario and sensitivity model development with the outputs and work back to the inputs. A four-week sequence works for most companies that already have a [three-statement model](https://treelife.in/finance/three-statement-financial-model-services/).

- **Week 1:** write one paragraph of narrative for each case, list twenty to thirty candidate drivers and run a one-way sensitivity on each to rank them.

- **Week 2:** fix the scenario layer at five to eight drivers, build the input sheet with one column per case and a Live column, and add the management action switch.

- **Week 3:** wire the selector, build the data table comparison with its tie-out check, and run the tornado, two-way grids and breakeven.

- **Week 4:** build the output tables by audience, add the assumptions register, run the checks sheet and fix the case label and version stamp in every header.

The same discipline applies to monthly forecasting, covered in rolling budget forecast implementation and cash flow forecasting. For the full build of a startup model, the parent guide is startup financial modelling guide.

## FAQs on Scenario and Sensitivity Model Development

**Q:** What is scenario and sensitivity model development? 
**A:** It is the process of adding structured cases and driver tests to a financial model. Cases describe coherent alternative futures and sensitivities rank which drivers matter. A reviewer should be able to switch cases, read the label and tie every figure to the engine.

**Q:** Can DCF be used for tax fair market value of unquoted shares in FY 2026-27? 
**A:** No, not under Rule 57 of the Income-tax Rules, 2026, which prescribes a formula for unquoted equity shares and carries no DCF option. DCF remains usable for commercial valuation and for pricing under Rule 21 of the FEMA (Non-debt Instruments) Rules, 2019. Label model outputs by purpose.

**Q:** How is scenario and sensitivity model development priced? 
**A:** Fees follow scope: entities, cases, output tables and the quality of existing data. A scoping call with the current model in hand is the fastest route to a fixed quote. A driver-based model costs less to extend than one rebuilt from the books.

**Q:** How long does scenario and sensitivity model development take? 
**A:** Usually two to three weeks when the model is already driver-based and the books are clean. It takes longer when drivers must be built from raw accounting data or several entities must be consolidated.

**Q:** Which documents are needed to start? 
**A:** The current model, the last two years of financial statements, the latest MIS pack, the cap table, the term sheet or sanction letter, and driver data such as customer cohorts, pipeline, headcount plan and debtor ageing.

**Q:** Does a FEMA valuation need scenarios? 
**A:** Rule 21 of the FEMA (Non-debt Instruments) Rules, 2019 requires an arm’s length, internationally accepted methodology, certified by the prescribed professional. It does not name scenarios, but a DCF on weighted cases needs the cases, weights and discount rate documented.

**Q:** How should co-founders agree the downside case? 
**A:** Agree the narratives before the numbers, then show each founder’s post-round stake under every case. A downside that moves the capital arrival date shows dilution effects that a base case hides.

**Q:** Does DPIIT recognition change the scenario model? 
**A:** Not the mechanics, but it adds a threshold to watch. The startup tax holiday formerly under Section 80-IAC now sits at Section 140 of the Income-tax Act, 2025, needs an Inter-Ministerial Board certificate on top of recognition, and carries its own turnover condition. Recognition itself now allows turnover up to ₹200 crore under G.S.R. 108(E) of 04/02/2026, as covered in the revised startup recognition framework. Model tax as a switchable driver, because the upside case can cross a turnover limit and switch the benefit off.

**Q:** What if the funding round does not close? 
**A:** Build a no-raise case using the capital arrival date driver. Set the raise to zero or move the date by two quarters, switch on the management action flag and read the cash exhaustion year.

**Q:** What do investors check in the scenario layer? 
**A:** That cases are narratives, not relabelled sensitivities, that the downside is shown before management action, that data tables tie to the engine, and that valuation does not rest almost entirely on terminal value.

**Q:** How do ESOPs and NRI founders enter the model? 
**A:** Treat pool top-ups as a driver and show the diluted founder stake by case. For NRI founders or non-resident investors, carry INR/USD as a driver and record the pricing basis against Rule 21 of the FEMA (Non-debt Instruments) Rules, 2019.

**Q:** When does Monte Carlo simulation make sense? 
**A:** When you have defensible probability distributions for each driver, which most growth-stage companies lack. Weak distributions give false precision, so cases plus a tornado ranking are easier to defend.

**Q:** When should we move from Excel to a planning platform? 
**A:** When several entities, many contributors or a slow monthly refresh make version control the bottleneck. A platform adds audit trail but does not fix unclear cases or drivers. The architecture here ports to either.

**Q:** Should scenario cases carry probability weights? 
**A:** Use weights when a single value is needed, such as for a negotiated round or a certifier’s report, and keep them as visible inputs. For a board discussion of risk, show cases side by side without weights, so the debate is about the narrative.

---

##### **Regulatory references**

- Income-tax Act, 2025, in force from 01/04/2026

- Income-tax Rules, 2026, Rule 57 (determination of fair market value), Notification No. 22/2026 dated 20/03/2026, effective 01/04/2026

- Section 56(2)(viib), Income-tax Act, 1961, omitted by the Finance (No. 2) Act, 2024 with effect from 01/04/2025

- Rule 21, Foreign Exchange Management (Non-debt Instruments) Rules, 2019, under the Foreign Exchange Management Act (FEMA), 1999

### Related posts:

- [What Is An Income Statement?](https://treelife.in/finance/what-is-an-income-statement/)

- [The Rise & Fall Of Indian IPO’s](https://treelife.in/finance/the-rise-fall-of-indian-ipo/)

- [Tyke’s CSOPs: Bridging Startups with Investors or Crossing Regulatory Boundaries?](https://treelife.in/finance/tykes-csops-bridging-startups-with-investors-or-crossing-regulatory-boundaries/)

- [Revised Valuation Rules for Angel Tax](https://treelife.in/finance/revised-valuation-rules-for-angel-tax/)

---

This is informational content from Treelife. For advice specific to your situation, contact support@treelife.in