Financial Model for New Product Launch – A Complete Guide

A financial model for new product launch answers one question before money is committed: does the product earn back its build cost, its launch spend and the revenue it takes from your existing range, within a cash position the business can carry? In India the answer also depends on GST credit timing, vendor payment rules and the tax regime the company has chosen. Most templates stop at a revenue curve and a break-even chart. This guide builds the model the way a board reads it: incremental, stage-gated and tied to cash.

How do you build a financial model for a new product launch?

Build it as an incremental model in five steps: define the decision and horizon, set volume, price and cost drivers in an assumption book, subtract revenue taken from existing products, add launch spend, capex and working capital to reach a monthly cash trough, then test payback, NPV and break-even under base, downside and upside cases. Release spend in stages against evidence.

What should a financial model for new product launch answer?

It should answer four things: how much cash the launch consumes before it recovers, when the product pays back, how much of its revenue is genuinely new, and which assumption breaks the case first. It is a decision model for an operating business, not a three-statement forecast of the whole company. Keep that boundary and the model stays small enough to audit.

The unit of analysis is the product, and the view is incremental. Incremental means the difference between the company with the product and the company without it. That difference includes revenue the product takes from your existing range, cost the existing team absorbs and cash the existing balance sheet carries. A model that reports only the gross revenue of the new product overstates the case.

The horizon follows product life, usually three to five years, with a support tail and a sunset case at the end. A short-horizon decision needs a detailed one-year view, and a long-horizon decision needs direction more than detail. Forecast adoption by quarter or year when there is no history, because monthly revenue forecasts add false precision. Build the cash schedule by month for the first 24 months, because the trough hides in annual columns.

Four boundaries keep a launch model small:

• It covers one product or one tightly linked bundle, not the company.
• It reads actuals of the existing line from the books, so cannibalisation is measured and not guessed.
• It stops at free cash flow after tax, with funding structure in a separate model.
• It carries a stage-gate schedule, so the output is a spend release plan and not only a forecast.

Table 1: Which planning model answers which question

Model	Question it answers	Horizon	Where Treelife covers it
Product launch model	Should we fund this product, and in what stages?	Product life, usually 3 to 5 years	This article
Startup three-statement model	What does the whole company look like under a strategy and funding path?	3 to 5 years	Financial modeling for startups
Revenue forecast model	What revenue will each stream deliver, by driver, cohort and pipeline?	12 to 24 months	Revenue forecast model development
Scenario and sensitivity model	Which cases and drivers move the company’s outcome?	Company plan horizon	Scenario and sensitivity model development
Funding requirement model	How much should the next round raise?	Milestone based	Startup funding requirement model
Working capital and cash flow model	How does cash move through receivables, inventory and payables?	24 months, monthly	Working capital and cash flow model preparation
Budget and rolling forecast	Are we spending to plan this year?	12 months	Budget forecasting for startups

The rest of this guide covers the first row only. Where a topic belongs to another row, the table above is the pointer.

Do you need a screening filter before the full model?

Yes. Run a five-check screen first. It takes hours, and it stops weak ideas before anyone builds a model. The checks are incremental contribution per unit, break-even volume, payback, reserve coverage and cannibalised share. Only ideas that screen green or amber justify the full launch model.

Table 2: Five-check screen before the full launch model

Check	Green	Amber	Red
Incremental contribution per unit	Positive after charging shared capacity	Positive only before charging shared capacity	Negative
Break-even volume	Reached inside Year 2 at base volume	Reached inside Year 3	Not reached inside product life
Payback	Inside half of product life	Inside product life	Beyond product life
Reserve coverage (cash the company can commit, divided by peak funding need)	1.5 times or more	1.0 to 1.5 times	Below 1.0 times
Cannibalised share (lost contribution as a share of gross contribution)	Under a quarter	A quarter to a half	Over half

These thresholds are illustrative. Set your own at the first gate. Two ambers or any red means shrink the launch or buy evidence before building.

Which inputs go into the assumption book?

A financial model for new product launch rests on seven input blocks: market and volume, price, unit cost, launch one-time spend, capital expenditure, working capital days and tax. Every cell carries an owner, a source and a date. A launch model fails when a number cannot be traced to a person who can defend it in the board meeting.

The seven blocks, in the order the model reads them:

• Market and volume: named segments, conversion path and the ceiling set by capacity, inventory or hiring.
• Price: list price net of GST, discount depth and billing cycle.
• Unit cost: variable cost per unit, payment gateway or logistics cost and support cost per customer.
• Launch one-time spend: build, regulatory approvals, launch marketing and channel set-up.
• Capital expenditure: equipment, tooling and licences, with the date of purchase.
• Working capital days: receivable, inventory and payable days by customer and vendor class.
• Tax: GST rate by HSN or SAC code, the company’s income tax regime and any holiday or exemption it holds.

Assign one owner to each block, because a launch cuts across functions:

• Product and engineering own the build estimate and timeline.
• Marketing owns launch spend and acquisition cost.
• Sales owns pipeline, conversion and price realisation.
• Operations owns unit cost, capacity and vendor terms.
• Finance owns price net of GST, tax, working capital and the model itself, and keeps one approved version.

How do you source assumptions when the product has no history?

Grade every assumption by the strength of its evidence, and let the base case rest only on the stronger grades. A paid pilot outranks a market report, and a founder estimate belongs in the upside case. Pull twelve months of the existing line from the books and GST returns, so that price, discounts and returns are real. Treelife’s note on what GST returns reveal that a P&L hides explains why the two should reconcile.

Table 3: Evidence ladder for launch assumptions

Evidence grade	Example source	Where the model uses it	Rule
Grade 1	Signed orders, paid pilots, letters of intent	Volume and price in the base case	Convert each into a dated cash event
Grade 2	Conversion of your own customer base on a comparable earlier upsell	Adoption curve	Use the same cohort window the earlier product used
Grade 3	Your own analogue product at a similar price point	Repeat rate and churn	Adjust for channel and ticket size
Grade 4	Published market data and market sizing (TAM, SAM, SOM)	Market ceiling only	Never the driver of base volume
Grade 5	Founder estimate	Upside case only	Record the owner and the date

When the base case needs grade five numbers to work, the model is a plan and not evidence. The first gate should then buy evidence, not build the product. Also check the share of the serviceable market that Year 3 volume implies. A base case that needs a large share is a plan and not evidence.

How do you forecast volume and cannibalisation for a launch?

Forecast volume bottom-up from named customer segments and a conversion path, then subtract the revenue each new sale takes from an existing product. The net figure is what the launch adds. A top-down assumption such as one per cent of a large market is not evidence and does not survive a board question.

Which volume method fits a launch with no history?

The method follows how the buyer arrives and the evidence you hold. Treelife’s revenue forecast model guide covers driver, cohort and pipeline methods in full. A launch with no history leans on three:

• Existing-base conversion: the share of current customers who adopt over time. It fits add-ons, tiers and features.
• Pipeline method: named accounts, multiplied by stage conversion and ticket size. It fits B2B products with identifiable buyers.
• Analogue curve: the monthly adoption curve of your own earlier launch, scaled to the new price and base.

Cap volume by supply whichever method you pick. Hiring, capacity, inventory and onboarding bandwidth set a ceiling that a demand curve ignores.

Revenue in the launch year is earned on average customers, not year-end customers. A module that ends Year 1 with 40 customers earned on about 20 if customers arrive evenly, so price average customers, net of churn. Using year-end customers doubles Year 1 revenue in that case.

How do you model cannibalisation of existing products?

Estimate the share of new-product buyers who would otherwise have bought, or kept, an existing product. Multiply that share by the existing product’s price and contribution margin to get lost contribution. Subtract it from the new product’s contribution. The result is incremental contribution, the only contribution that justifies the launch.

The arithmetic is three lines:

• Gross contribution equals new customers, multiplied by price, multiplied by one minus the variable cost share.
• Lost contribution equals new customers, multiplied by the cannibalised share, multiplied by the old price, multiplied by one minus the old variable cost share.
• Incremental contribution equals gross contribution minus lost contribution.

Measure the cannibalised share from your sales history where an earlier tier launch exists. Otherwise run a pilot in which buyers are asked what they would have bought instead.

Cannibalisation can also run the other way: a product that raises retention of the existing range adds contribution, on its own line with its own evidence.

Shared capacity is not free. If the launch uses an engineer, a production line, shelf space or a sales team that earns revenue elsewhere, charge that opportunity cost to the launch. Where the capacity is idle today, charge only the incremental cost of hiring or expanding it.

How should launch costs and unit economics be built?

In a financial model for new product launch, split cost into five behaviours: variable cost per unit, one-time build, one-time launch spend, capital expenditure and fixed run cost. Each behaves differently in cash and in the accounts. Mixing them hides contribution margin, which is the number that decides whether growth adds value or only adds spend.

Price in the model is net of GST. Tax collected is a liability to the government and not revenue, so a model that carries GST-inclusive prices overstates revenue by the tax rate. The principal GST slabs are now 5% and 18%, with 40% for select demerit goods, effective 22/09/2025 under Notification No. 9/2025-Central Tax (Rate) dated 17/09/2025. Special rates and exemptions still exist, so check the HSN or SAC code of the new product. A new product can land in a different slab from your existing range.

The accounting treatment changes the profit line but not the cash line. Under Ind AS 38 (Intangible Assets), research cost is expensed (paragraph 54). Development cost is capitalised only when six criteria are met, including technical feasibility and intention to complete (paragraph 57). Advertising and promotional spend is always expensed (paragraph 69). The capitalisation call moves launch-year EBITDA by the full development spend while cash stays unchanged, so show EBITDA and cash side by side.

Launch spend does not stop at launch. Support, product upkeep and acquisition continue every month after the first sale, so start run cost in the month of release and not after the build ends. Build each one-time cost from a driver and not a lump sum:

• Development: sprints multiplied by cost per sprint.
• Launch marketing: target leads multiplied by cost per lead.
• Tooling: number of SKUs multiplied by tooling cost per SKU.
• Onboarding: customers per implementer multiplied by cost per implementer.

Table 4: Launch cost lines and how the model treats them

Cost line	Behaviour	Model treatment	Accounting or tax reference
Variable cost per unit	Scales with volume	Driver multiplied by volume, net of recoverable GST	Section 16(2), CGST Act 2017
One-time build or development	Fixed and front-loaded	Cash in build months; capitalise only if criteria are met	Ind AS 38, paragraphs 54 and 57
Launch marketing	Fixed, then tied to acquisition	Cash as incurred; expensed in the P&L	Ind AS 38, paragraph 69
Equipment and tooling	Capital and lumpy	Depreciate over useful life	Schedule II, Companies Act 2013
Fixed run cost and new hires	Step-fixed	Hire date drives cost; show share-based pay separately	Ind AS 102
Payables to micro and small vendors	Cash timing	Pay within the MSMED Act limit	Section 37(2)(g), Income-tax Act 2025

Acquisition cost deserves its own row in the model. Acquisition cost per new customer, divided by monthly contribution per customer, gives CAC payback in months. In a launch, acquisition cost is highest and contribution lowest in the first months, so CAC payback is the metric that explains the early cash trough. Compute lifetime value on contribution and not revenue, and set the LTV to CAC floor from your own cohorts. For subscription products, Treelife’s SaaS CAC, LTV and churn guide covers the build.

How much cash does a product launch consume before it pays back?

A financial model for new product launch should size peak cash need as the lowest cumulative cash position across the launch months, not the first-year loss. It includes build cost, launch spend, working capital and GST credit that is paid out before it is recovered. Show this trough by month, so the board funds the launch from cash it holds and not cash it expects.

Receivable, inventory and payable day mechanics sit in Treelife’s working capital and cash flow model guide. A launch model borrows that schedule and adds the items below.

Which India-specific cash items change the trough?

Five items shift the trough most often. Each needs a driver in the model and not a plug.

Table 5: Cash timing items that shift the launch trough

Item	What to model	Timing risk	Reference
Input GST credit on launch spend	Credit arises only when the supplier’s invoice reaches your return	Cash leaves at payment, credit offsets output tax later	Section 16(2), CGST Act 2017
Accumulated credit under inverted duty	Output rate below input rate leaves credit unused	Refund was limited to credit on inputs until 31/10/2026; 90% provisional refund and service credit availed from 01/11/2026 now count	Sections 54(3) and 54(6), CGST Act 2017; Rule 89(5), CGST Rules 2017
Payments to micro and small vendors	Pay within the agreed term, never beyond the statutory limit	Deduction deferred to the year of payment if paid late	Section 37(2)(g), Income-tax Act 2025; Section 15, MSMED Act 2006
Receivables from corporate customers	Days outstanding by customer class, net of tax deducted at source	Deducted tax stays a credit until the return is filed	Customer contract terms
Inventory or capacity pre-commitment	Purchase orders placed before the first sale	Cash locked before demand is proven	Release only at the launch gate

Rate rationalisation from 22/09/2025 can leave credit stuck for a new product whose output rate sits below its input rate. Until 31/10/2026, Section 54(3)(ii) of the CGST Act 2017 and Rule 89(5) limited refund to credit on inputs. From 01/11/2026, the 90% provisional refund under Section 54(6) extends to inverted duty claims (Section 155, Finance Act 2026; Notification No. 03/2026-Central Tax dated 26/10/2026). Notification No. 05/2026-Central Tax (G.S.R. 919(E)) dated 26/10/2026 substitutes Rule 89(5), so Net ITC includes eligible input services availed on or after 01/11/2026 and capital goods credit spread over 60 months from 01/04/2027, following the 57th GST Council recommendation of 08/10/2026. The Finance Act 2026 amendments in force from 01/11/2026 cover Sections 15, 34 and 54(6), and the Council’s recommended change to the proviso to Section 54(3)(ii) is not among them. Model refund of service credit as delayed cash in the base case and as nil in the downside case until that proviso is amended. Treelife’s D2C guide works the credit arithmetic in full.

Tax on launch profit follows the company’s regime, an effective 25.17% under Section 200 of the Income-tax Act 2025 (earlier Section 115BAA of the 1961 Act) for a company that opts in. The launch-specific point is the shield. Early launch losses cut tax only where the company has taxable profit elsewhere in the same year.

If new equity funds the trough, carry the number into the funding requirement model and the burn rate and runway calculation, and run dilution there.

How do you decide go or no-go for a product launch?

A financial model for new product launch should set the rule before the numbers arrive: payback inside the product’s expected life, a cash trough the company can fund from existing cash, and incremental NPV above zero at a discount rate that reflects launch risk. Then release spend in gates, each with a kill trigger written in numbers.

Six outputs carry the decision:

• Payback period: months until cumulative incremental cash turns positive.
• Break-even volume: fixed launch costs divided by incremental contribution per unit, counted cumulatively and not month by month.
• Incremental NPV and IRR: discounted incremental free cash flow over product life, with terminal value kept out of the base case.
• Peak funding need: the lowest point of cumulative cash, by month.
• Reserve coverage: cash the company can commit, divided by peak funding need. Below one, the launch cannot be funded from own cash.
• CAC payback: months to recover acquisition cost from contribution.

For the discount rate, use the company’s cost of capital plus a premium for launch risk. Record the premium and test two values.

How do stage gates control launch spend?

A stage gate releases a defined amount of spend only after named evidence arrives. It converts the model from a forecast into a control. Write the trigger numbers into a control tab at the first gate, while nobody is attached to the result. A kill criterion drafted after a disappointing pilot becomes a negotiation.

Table 6: A four-gate launch schedule

Gate	Evidence required to pass	Spend released	Kill or redesign trigger
Gate 0, concept	Interviews and a priced offer tested with named accounts	Discovery budget only	No buyer commits at the modelled price
Gate 1, pilot	Paid pilots meeting modelled conversion and cannibalised share	Build to minimum scope	Conversion or repeat below the downside case
Gate 2, launch	Unit cost confirmed by vendor quotes; channel CAC tested	Launch marketing, inventory or capacity	CAC payback beyond the limit set at Gate 0
Gate 3, scale	Actuals inside the base case band for a full cohort cycle	Scale spend and hiring	Contribution margin below the floor set at Gate 0

What does a financial model for new product launch look like in numbers?

An incremental launch model shows three things a gross model hides: payback is longer, peak cash need is the build cost plus the first-year loss, and value is lower once cannibalised contribution is removed. The example below is illustrative: a B2B software company adds a premium module sold to its existing customers. All figures are assumptions in ₹ lakh, pre-tax and before working capital.

The assumptions:

• Build cost of ₹60 lakh in the build year, before the first sale.
• Price of ₹1.2 lakh per customer per year, net of GST.
• Variable cost of 20% of revenue for both products.
• Fixed run cost of ₹30 lakh a year, and launch marketing of ₹25 lakh in Year 1 and ₹20 lakh a year after.
• Average paying customers of 20, 80, 140 and 200 across Years 1 to 4.
• One in four new buyers would otherwise have bought a ₹0.6 lakh plan.

Table 7: Illustrative incremental cash view of a module launch (₹ lakh, pre-tax)

Line	Year 1	Year 2	Year 3	Year 4
Gross revenue of new module	24.0	96.0	168.0	240.0
Gross contribution (80% of revenue)	19.2	76.8	134.4	192.0
Contribution lost to cannibalisation	2.4	9.6	16.8	24.0
Incremental contribution	16.8	67.2	117.6	168.0
Fixed run cost and launch marketing	55.0	50.0	50.0	50.0
Incremental cash profit	(38.2)	17.2	67.6	118.0
Cumulative cash after the ₹60 lakh build	(98.2)	(81.0)	(13.4)	104.6

Three readings follow from the table. First, the module does not pay back inside three years on the incremental view: cumulative cash is still ₹13.4 lakh negative at the end of Year 3, and payback lands at about 3.1 years. Second, the gross view that ignores cannibalisation shows payback at about 2.8 years, which flatters the case by roughly four months. Third, peak cash need is ₹98.2 lakh, the build cost plus the Year 1 loss, and not the ₹38.2 lakh Year 1 loss alone.

Two further outputs complete the read. Each average paying customer adds ₹0.84 lakh of incremental contribution a year (₹1.2 lakh at an 80% margin, less ₹0.12 lakh lost to cannibalisation per customer), so ₹50 lakh of annual run cost and marketing needs about 60 average paying customers, a line the module crosses during Year 2. At a 15% discount rate, an assumption, and no terminal value, incremental NPV over the four years is about ₹31.7 lakh and IRR about 26.5%. The gross view shows NPV of about ₹65.8 lakh and IRR of about 37.7%, so ignoring cannibalisation roughly doubles the apparent value.

Run through the screen in Table 2, the module is green on contribution, break-even and cannibalised share, and amber on payback, at about 3.1 years against a four-year life.

Financial model for new product launch: cumulative cash, gross view against incremental view

Which scenarios and sensitivities should the model run?

Run three cases and one reverse test. Base, downside and upside cases flex the same driver set. The reverse test asks what must be true for payback to fall inside product life, then checks whether that is believable. Sensitivities show which driver deserves evidence first.

Treelife’s scenario and sensitivity guide covers case construction and output tables. For a launch, flex these drivers one at a time before combining them:

• Price and discount depth
• Adoption speed in the first six months
• Cannibalised share
• Acquisition cost per customer
• Launch delay in months
• Customer payment days

In the example, doubling the cannibalised share from 25% to 50% pushes payback from about 3.1 to about 3.4 years and deepens the trough to ₹100.6 lakh.

How do you track launch actuals against the model?

Tag every launch rupee at source, so actuals compare with the model line by line. Give the product its own item code, cost centre and revenue head, then read monthly variance against the gate triggers and replace forecast months with actuals. Without tagging, launch results disappear into company totals and the kill criteria cannot be tested.

Four habits make this work:

• Placeholder before launch: create the product as a placeholder item in the plan before the SKU exists, then map it to the live item code so variance stays traceable.
• One tagging rule: item code with HSN or SAC, a launch cost centre, and a tag on every launch hire and vendor invoice.
• A monthly variance pack: a one-page KPI summary of volume, price, cannibalised share, acquisition cost, cash trough and kill trigger status, in that order.
• A refresh rule: replace the oldest forecast month with actuals, recompute payback and the trough, and let each gate decide on actuals and not on the original plan.

The monthly pack sits naturally inside a MIS and budgeting routine.

Common mistakes that cost founders time and money

Five mistakes distort most launch decisions: treating gross revenue as incremental, sizing volume from market share, funding the first-year loss, ignoring GST and vendor payment timing, and releasing the full budget at once.

• Treating gross revenue as incremental. It happens because the new product’s P&L looks healthy on its own. Subtract lost contribution. In the example above, skipping this step understates payback by about four months.
• Forecasting volume as a share of the market. It happens because the number is easy to defend with a report. Build from named segments and grade the evidence, as in Table 3.
• Funding the first-year loss and not the trough. It happens because the first-year loss is the smaller number. Fund the lowest point of cumulative cash. In the example that is ₹98.2 lakh against a ₹38.2 lakh first-year loss.
• Ignoring GST credit and vendor payment timing. It happens because the P&L hides both. A payment to a micro or small vendor beyond the Section 15 MSMED Act limit defers the deduction for that sum to the year of actual payment under Section 37(2)(g) of the Income-tax Act 2025. Model payment days by vendor class.
• Releasing the full budget on approval. It happens because momentum feels like progress. Release spend by gate and write the kill triggers before the pilot, as in Table 6.

Practitioner note

In the financial model for new product launch engagements we have run at Treelife, the first version usually fails on the same three points. It prices in GST-inclusive terms, it treats input credit as cash on the day of purchase, and it reports the gross revenue of the new product as the launch’s contribution. Section 16(2) of the CGST Act 2017 gives credit only when the supplier’s invoice reaches your return, so credit arrives after cash has left, and for a launch with heavy services spend that gap is the trough.

The pattern that only live work shows is the capitalisation conversation. Finance teams decide late whether development cost is capitalised under paragraph 57 of Ind AS 38, and the answer moves launch-year EBITDA by the entire build cost. Investors read EBITDA, boards read cash, and a model that shows only one invites an argument in the meeting. We build both views from the same cost tab and agree the capitalisation policy with the statutory auditor before the model is shared.

The second pattern is cannibalisation measured too late. Founders often learn the cannibalised share from the first cohort’s invoices, after the price card is public. A pilot that asks buyers what they would otherwise have bought costs little and moves the case more than any other input. Run it at the pilot gate, before build spend is released.

FAQs on Financial Model for New Product Launch

Q: How do launch losses and launch spend affect the tax line in the model?
A: Apply the tax effect only where the company has taxable profit in the same year. A domestic company that opts for Section 200 of the Income-tax Act 2025 (earlier Section 115BAA of the 1961 Act) pays 22% plus 10% surcharge and 4% cess, an effective 25.17%. In a loss-making company, show the launch loss without a tax shield and keep the carried-forward loss in the company model.

Q: Does a launch qualify for the startup tax holiday?
A: Only if the entity already holds the holiday and the product sits inside its certified business. Section 140 of the Income-tax Act 2025 (earlier Section 80-IAC) gives a 100% deduction for three consecutive years out of the first ten, for DPIIT-recognised companies and LLPs with Inter-Ministerial Board certification, incorporated before 01/04/2030. Model the deduction as a switch that turns on only for in-scope profit.

Q: What does a financial model for new product launch cost to build?
A: The cost follows scope: number of products, depth of cannibalisation analysis and whether monthly tracking is included. Engagements are usually a fixed build fee plus a retainer for monthly actuals.

Q: How long does a financial model for new product launch take to build?

A: The build is short once inputs arrive. Pace is set by how fast the owners of price, cost, volume and vendor terms return evidence.

Q: Which documents does the model need?
A: Eight inputs close most gaps:

• Twelve months of sales and GST returns for the existing line
• Rate cards and discount history
• Vendor quotes for build, equipment and services
• Hiring plan with start dates
• Pilot or pipeline records
• Customer and vendor payment terms
• Latest tax computation
• Current cap table and board approval limits

Q: How do I model a launch first sold to overseas customers?
A: Model export sales as zero-rated supplies under Section 16 of the IGST Act 2017, shipped under a Letter of Undertaking (Rule 96A, CGST Rules 2017) with refund of unused credit under Section 54 of the CGST Act 2017. Add foreign exchange rate, realisation lag and the rules of the Foreign Exchange Management Act (FEMA) 1999 as separate drivers. If the launch needs a foreign subsidiary, the investment sits under the Foreign Exchange Management (Overseas Investment) Rules, 2022. Treelife’s D2C guide works the export cash mechanics.

Q: How should co-founders agree who can approve launch spend?
A: Write approval limits by gate into the shareholders’ agreement or board delegation. Investment of company funds is a power reserved to the board under Section 179(3) of the Companies Act 2013. The model’s control tab then doubles as the board paper.

Q: What happens to the spend if the launch stops at a gate?
A: Spend already incurred is sunk, and the model should show the kill-case cash separately. Capitalised development cost is tested for impairment under Ind AS 36, and unsold inventory or committed vendor orders become the real exposure. This is why inventory and capacity commitments are released only at the launch gate.

Q: What do investors check in a launch model?
A: They check that actuals of the existing line reconcile to the books and GST returns, that cannibalisation is measured, and that the launch spend ties to use of funds. They also check version control. Treelife’s investor pitch projections guide covers how the same discipline applies to a round.

Q: How should launch hires and their ESOPs enter the model?
A: Hire dates drive cost. Show share-based payment expense separately, because under Ind AS 102 it is non-cash and does not enter the trough. Pool size affects dilution, which belongs in the funding requirement model and not in the launch model.

Q: Can an NRI founder or overseas promoter fund the launch?
A: Yes, through an instrument and route that comply with the Foreign Exchange Management (Non-debt Instruments) Rules, 2019 as amended in 2026, including the entry route, sectoral cap and pricing rules, and prior approval where a land-border country is involved. Place the inflow in the model after allotment and reporting, not at board approval, because the cash arrives only after the compliance steps close.

Q: How do I model a launch delay?
A: Shift the revenue curve right and keep costs running. In the worked example, Year 1 run cost and marketing of ₹55 lakh is about ₹4.6 lakh a month, so a three-month delay deepens the trough by about ₹13.8 lakh. Run delays of three and six months as named scenarios, because timing moves the trough even when price and volume hold.

Q: How do I model the end of a product’s life?
A: Add a decline curve, a support tail and a sunset month, and move migrating customers back to the existing plan. Review the useful life of capitalised development cost, because amortisation under Ind AS 38 runs over useful life and an early sunset can trigger an impairment test under Ind AS 36. Carry a sunset case in the downside scenario.

Build the launch model before the launch budget

A financial model for new product launch earns its place when it changes a decision: how much cash to hold, what to release at each gate and when to stop. Build it incrementally, price it net of GST, show the trough by month and write the kill triggers down first.

---

Regulatory references

• Income-tax Act 2025, Section 200 (tax on income of certain domestic companies; earlier Section 115BAA, Income-tax Act 1961)
• Income-tax Act 2025, Section 37(2)(g) (payments to micro and small enterprises; earlier Section 43B(h), Income-tax Act 1961)
• Income-tax Act 2025, Section 140 (startup deduction; earlier Section 80-IAC, Income-tax Act 1961)
• Micro, Small and Medium Enterprises Development Act 2006, Section 15
• Central Goods and Services Tax Act 2017, Sections 16(2), 54(3) and 54(6)
• Central Goods and Services Tax Rules 2017, Rules 89(5) and 96A
• Integrated Goods and Services Tax Act 2017, Section 16
• CBIC Notification No. 9/2025-Central Tax (Rate) dated 17/09/2025, effective 22/09/2025
• Finance Act 2026, Section 155 (Section 54(6), CGST Act 2017); CBIC Notification No. 03/2026-Central Tax dated 26/10/2026, effective 01/11/2026