Startup Valuation methods in India: DCF, Comparables and Registered valuer Reports

Authored By - Hemanth Aligingi, Senior Executive at Outsource360

In June 2026 a Bengaluru-headquartered cloud-kitchen company did something that should interest any founder who has ever argued about a startup valuation: it shelved its own listing. SEBI had cleared the offer back in October 2025. The company was seeking a valuation of roughly Rs 4,000 crore, and the offer itself was an Rs 800 crore fresh issue plus an offer for sale of up to 4.85 crore equity shares.

Then came the roadshows. Institutional buyers looked at the number, declined to meet it, and the IPO went on hold, as Inc42 reported on 8 June 2026 and Outlook Money corroborated.

The financials underneath the disagreement were not bad, exactly. Total income for FY25 came in at Rs 775.49 crore, up 22 per cent from Rs 635.08 crore the year before, and the loss narrowed slightly to Rs 169.96 crore from Rs 172.61 crore in FY24. Growing fast, still burning: a description that fits a large share of the Indian startup universe. A source quoted in the reporting put the buy-side view plainly, saying that for a loss-making company in that market a good premium was "hard to get".

The company had been priced for years by private investors who liked the story. The public market looked at the same business through a different lens and produced a different answer. Nobody disputed the revenue figure. What they disputed was what the revenue figure was worth.

So the company had a number. Its bankers had a number. The buyers had a different number, and the buyers' number won.

That is not an IPO phenomenon. It is what happens at every layer below one, and it is the reason "what is my startup worth" has no single answer.

A startup does not have a valuation. It has several, produced by different methods, for different audiences, on different dates, each carrying a different legal consequence.

The number on your term sheet is a negotiated price. A registered valuer's report carries a statutory determination instead. What a foreign investor's compliance team files with its banker is a third thing again.

None of them is more real than the others. They are answers to different questions, and confusing one for another is where founders get hurt.

Most founders reading this will never file a DRHP. You will still hit four or five separate moments where somebody has to put a number on your company and sign their name under it: a priced round, a non-resident joining the cap table, an ESOP grant, a founder secondary, a preferential allotment to a new investor. Different statutes govern each one. Different methods are permitted, different professionals may certify them, and a report scoped for one purpose will not carry another.

And since 1 April 2026, when the Income-tax Act, 2025 and the Income-tax Rules, 2026 came into force, the machinery behind two of those moments has changed in ways most published guidance has not caught up with.

Startup valuation methods in India fall into two groups. Qualitative methods (Berkus, scorecard, risk factor summation) price pre-revenue companies on team, product and market. Quantitative methods (discounted cash flow, comparable company analysis, venture capital method) price revenue-stage companies on cash flows and multiples. Which method a valuation report may carry depends on the statute behind it.

Everything below is India-specific: Indian statutes, Indian regulators, Indian filing deadlines. The income-tax position in particular changed on 1 April 2026. The sections that follow separate the methods investors reason with from the methods Indian law will actually accept inside a signed report.

On this page

1. What a startup valuation actually is, and why there is never only one number

   
   

   • Pre-money, post-money and the arithmetic founders get wrong

   • Why a startup resists valuation in a way a going concern does not

   • One holder, one stroke of a pen, a 95 per cent cut

     
     

2. The nine startup valuation methods used in India

   
   

   • Qualitative methods for pre-revenue companies

   • Quantitative methods: DCF, comparables and precedent transactions

   • The venture capital method and the cost-to-duplicate floor

   • Net asset value, the method Indian tax law now prescribes

     
     

3. Discounted cash flow for an Indian startup

   
   

   • The four inputs, and which one actually moves the answer

   • Why an Indian discount rate is not a US discount rate

   • Terminal value, where most of the number hides

   • Where a DCF stops being useful

     
     

4. Comparable company analysis, multiples and precedent transactions

   
   

   • Building a comparable set when the listed comps are orders of magnitude bigger

   • Which multiple suits which business model

   • Precedent transactions, and why Indian deal data is thin

     
     

5. Which method fits which stage

   
   

   • Pre-revenue and seed

   • Series A to Series C

   • Growth stage, secondaries and the pre-IPO round

     
     

6. The four valuation reports an Indian startup actually needs

   
   

   • Two registered-valuer regimes, and why being one does not make you the other

   • The events that need no report at all

   • How to brief a valuer so the report survives its purpose

     
     

7. Income-tax valuation after 1 April 2026: Rule 11UA is now Rule 57

   
   

   • What Rule 57 actually says about unquoted equity shares

   • What disappeared on 1 April 2026

   • A fresh share issue now triggers no income-tax valuation in the company's hands

   • Where the obligation survives: a transfer under section 79, a receipt under section 92(2)(m)

   • Angel tax: what the sunset clause says, and what practitioners still disagree about

     
     

8. FEMA pricing when a non-resident is on the cap table

   
   

   • Floor price, ceiling price, and which direction you are going

   • Who may certify a FEMA valuation

   • FC-GPR and the thirty-day clock

   • Can one report serve both FEMA and income tax?

     
     

9. The Companies Act layer: section 247 and the registered valuer

   
   

   • Who may be appointed, and by whom

   • Preferential allotment under section 62(1)(c) and Rule 13

   • What a registered valuer's report has to contain

   • What happens if the valuer is not registered

     
     

10. ESOP valuation: the one place a merchant banker is now mandatory

    
    

    • Rule 15 and the fair market value at exercise

    • The 180-day window, and how to use it

    • Why the ESOP price sits below the round price, and how to defend the gap

      
      

11. What the tax tribunals have held about DCF, and whether it still matters

    
    

    • The principle: the choice of method is the assessee's

    • Projections against actuals, and the hindsight problem

    • Does the case law survive Rule 57?

      
      

12. What a valuation report costs, and how long it takes

    
    

    • The statutory fee cap nobody cites

    • What the market actually charges, and why the published figures disagree

    • What drives the number, and the turnaround to plan for

      
      

13. When the valuation number and the round price disagree

    
    

    • The round price is above fair value

    • The round price is below fair value, which is now the dangerous direction

    • The valuer's number comes in under the term sheet

    • The report goes stale before the board meets

      
      

14. If your holding company is in Delaware: 409A alongside Rule 57

    
    

    • Two entities, two standards, two numbers

    • What a flip does to the Indian entity's valuation obligations

      
      

15. Common valuation mistakes, and what a down round does to them

    
    

    • Treating the FEMA number and the tax number as interchangeable

    • Running on a report that a material event has overtaken

    • Modelling a round without the option pool

    • What a down round does to your last registered valuer report

      
      

16. Frequently asked questions

    
    

What a startup valuation actually is, and why there is never only one number

Ask ten people in a funding conversation what the company is worth and you get ten defensible answers. That is not sloppiness. A valuation is an opinion about the future produced by a method, and the method is chosen by whoever needs the answer. Change the purpose, change the method, change the number.

So the first useful question is never "what is my startup worth". It is: worth to whom, for what, and who has to sign it?

Pre-money, post-money and the arithmetic founders get wrong

Pre-money is what the business is agreed to be worth before the new money lands. Post-money is pre-money plus the cheque. Take a worked example, with figures chosen for the arithmetic rather than drawn from any real round: raise Rs 5 crore at a Rs 20 crore pre-money and the investor owns 20 per cent of a Rs 25 crore company. That part rarely goes wrong.

The option pool is where it does. Indian term sheets commonly require the ESOP pool to be created or topped up out of the pre-money, so founders fund it alone and the investor's percentage is computed after the pool exists. Carry the same illustrative figures forward: a Rs 20 crore pre-money with a 10 per cent pool carved in is not a Rs 20 crore pre-money in any sense a founder would recognise. Call it Rs 18 crore of founder value plus Rs 2 crore of pool, shown as one figure.

This is where most founders go wrong: comparing two term sheets on headline pre-money alone. Model both fully diluted, and read the round as one connected chain rather than a stack of separate documents.

Why a startup resists valuation in a way a going concern does not

Start with the easy case. Valuing a forty-year-old auto-components manufacturer is mostly arithmetic: an earnings history, a stable margin, a balance sheet full of saleable assets, listed peers doing much the same thing. Almost none of that exists for a five-year-old company.

A Series A SaaS business in Bengaluru typically has eighteen months of revenue, negative operating cash flow, and a balance sheet whose largest real asset is a team it does not record. Its listed "peers" are orders of magnitude larger. Most of its value sits in optionality: the honest word for the possibility that the current growth rate continues long enough to matter.

What's underappreciated is that every method below is a structured way of arguing about that probability. Some argue in the open. Others bury it inside a discount rate and hope nobody asks.

One holder, one stroke of a pen, a 95 per cent cut

Here's the thing about a headline valuation: it is an opinion held by whoever last wrote a cheque, and somebody who did not can revise it.

Between 2022 and 2024, US institutional holders of minority stakes in Indian startups repriced several of them unilaterally, with no new round. One asset manager cut its implied valuation of an Indian edtech company from about USD 22 billion to roughly USD 1 billion, a cut it made as at the end of October 2023 and disclosed in January 2024, a 95 per cent reduction reported by TechCrunch, and wrote the holding to zero five months later. In the same window, US holders cut an Indian food-delivery company from USD 10.7 billion to USD 8 billion, then to USD 5.5 billion four months later. A mobility company was cut by 35 per cent.

None of those markdowns came with a published basis. Same companies, same weeks, valuations differing by an order of magnitude. Keep that in mind: the number you quote in a pitch and the number a professional signs come from different processes, and only one carries consequences when it proves wrong.

The nine startup valuation methods used in India

Nine methods do most of the work in Indian startup valuation. They sort into three families: qualitative scoring models for companies with no revenue, market-based methods that price you against somebody else, and cash-flow or asset-based methods that price you against your own books.

Most guides stop there, listing all nine as interchangeable options on a menu. They are not, and the fourth column below is why. Five of them cannot appear in an Indian statutory valuation report at all, because no statute contemplates them. Investors use them constantly. A valuer cannot sign them.

Which of the nine actually survives a filing?

Method	What it needs	Best-fit stage	Can a statutory report carry it?
Berkus	A product, a team, a market	Pre-revenue	No. Investor tool only
Scorecard	Average pre-money for regional seed deals	Pre-revenue	No. Investor tool only
Risk factor summation	Risk categories scored up or down	Pre-revenue	No. Investor tool only
Cost to duplicate	Spend to date on product and IP	Pre-product	No. Investor tool only
Venture capital method	An exit assumption and a target return	Seed to Series B	No. Investor tool only
Comparable company analysis	Peer companies and a defensible multiple	Series A onward	FEMA yes, an internationally accepted methodology
Precedent transactions	Disclosed deal terms in the same sector	Series B onward	FEMA yes, where deal data exists
Discounted cash flow	A forecast, a discount rate, a terminal value	Series A onward	FEMA yes. Income tax no for unquoted equity since 1 April 2026
Net asset value (book value)	An audited balance sheet on the valuation date	Any stage	Yes. The only method Rule 57 row 4 permits for unquoted equity

Qualitative methods for pre-revenue companies

Three methods, one shared premise: with no revenue to discount, you score what actually exists.

Berkus does not value a business. It values five things a pre-revenue business might have, capping each: a sound idea, a working prototype, a quality management team, strategic relationships, and evidence of early sales. Add them up and you get a ceiling, not an estimate. The cap exists to stop a founder pricing an idea like a company.

Scorecard works by comparison. It takes an average pre-money for recently funded companies at the same stage and region, then moves it up or down across weighted factors: team strength, size of opportunity, product and technology, competitive environment, sales channels, and whether further investment is needed. A founder can run it in a spreadsheet (which is both its appeal and its weakness).

Risk factor summation is the most pessimistic of the three. It starts from a baseline for comparable companies and walks through roughly a dozen risk categories (management, stage, regulation, manufacturing, sales, funding, competition, technology, litigation, international, reputation, exit), scoring each as positive, neutral or negative.

So why can none of the three sign a filing? A registered valuer working under the Companies Act applies recognised valuation standards, and a professional certifying a price under FEMA has to apply an internationally accepted pricing methodology on an arm's length basis. "The management team scored a plus one" is neither of those things.

That does not make the three methods useless. It makes them investor-facing, which is a different job entirely.

Quantitative methods: DCF, comparables and precedent transactions

Discounted cash flow projects the free cash the business will generate over a forecast horizon, adds a terminal value for everything after it, and discounts the stream back to today at a rate reflecting the risk of not receiving it. Honest, in the sense that it shows its working. Also the easiest method on this list to abuse, because every input is an assumption you chose.

Comparable company analysis asks what the market currently pays for businesses like yours, as a multiple of revenue, ARR, GMV or EBITDA, and applies it to your figure. Fast, market-anchored, and it collapses the moment the comparable set is bad. For most Indian startups, it is bad.

Precedent transaction analysis does the same using completed deals rather than trading prices. Because an acquirer usually pays a control premium, precedent multiples sit above trading multiples for similar companies, which is exactly why they get quoted selectively in pitch decks.

All three can support a FEMA certificate, since the pricing test there is any internationally accepted pricing methodology applied on an arm's length basis. DCF and comparables each get a section below, because the way they break is more instructive than the way they work.

The venture capital method and the cost-to-duplicate floor

So how does an investor actually arrive at a number? The venture capital method is how a seed or Series A investor reasons, whatever the term sheet says.

Start at the end: estimate what the company could be sold for or listed at in five to seven years, usually by applying an exit multiple to forecast revenue. Divide that exit value by the return the fund needs on this cheque. That gives a post-money today; subtract the investment for a pre-money.

Two things follow, and both matter in a negotiation. The required return is a property of the fund, not your company, so two investors with identical views can legitimately offer different prices. And because the method runs backwards from an exit, an investor who cannot picture a buyer for you will not produce a number at all.

Cost to duplicate is the opposite. What would a competent team spend today to rebuild what you have built: engineering hours, IP, data, approvals? It produces a floor rather than a value, and ignores brand, customer relationships and growth rate.

Frankly, this gets overlooked. Cost to duplicate is the most useful method for a founder who feels lowballed at pre-seed, because it puts a defensible number on work already done rather than on a story about work to come.

Net asset value, the method Indian tax law now prescribes

Net asset value, or NAV, is arithmetic, not judgment. Take the assets on an audited balance sheet, strip out what the rule tells you to strip out, subtract the liabilities the rule recognises, divide by the paid-up equity. No forecast, no discount rate, no negotiation.

For an early-stage company the result is usually a fraction of any commercial valuation, often close to nothing. That is not a flaw in the method. It is what happens when you measure a business whose entire value is future cash with a tool designed to measure present assets.

Worth flagging early: since 1 April 2026 this is the method Indian income-tax law reaches for on unquoted equity shares, and the income-tax section below works through exactly how it applies and where it bites.

Discounted cash flow for an Indian startup

A DCF is the method founders most want to use and most often use badly. The appeal is obvious: it values your company on your own plan rather than somebody else's multiple, and if the plan is good the number is high.

The catch? It shows up the moment you have built one. Two of its four inputs are almost pure assumption, so a model can move several-fold on changes that each look reasonable. Where does that leave a founder who genuinely believes the plan?

The four inputs, and which one actually moves the answer

So which input deserves the stress test? Input one is the cash-flow forecast, typically five years: revenue, costs, working capital, capital expenditure, tax. Input two is the discount rate, and input three is the terminal value, everything beyond the forecast horizon. Input four, the one founders forget, is future dilution, because the equity value you are solving for belongs to shares that do not exist yet.

Sensitivity-test it and the ranking is usually the same: the forecast years matter least on their own, the discount rate a great deal, the terminal value most. Move the discount rate two percentage points on a five-year model and the answer swings hard. Change year-four revenue by 10 per cent in isolation and it barely registers, though in a live model that change flows into the terminal year as well, which is where it bites.

The real work in a DCF is not building the spreadsheet. It is defending the two assumptions nobody can verify. Somebody has to own that model and its history, and at seed and Series A that person often does not exist in-house (which is much of what a virtual CFO actually does for a startup).

Why an Indian discount rate is not a US discount rate

Here's what that actually looks like when a valuer builds one. The discount rate is the return an investor would require to bear this company's risk, and it is assembled from components: it starts with a risk-free rate (an Indian government security yield, not a US Treasury yield), adds an equity risk premium, then increments for business size and company-specific risk such as customer concentration or single-founder dependency.

Every one of those components is larger for an Indian early-stage company than a comparable US one. Using a US cost of capital on an Indian cash-flow stream is not conservative. It is the wrong instrument, and produces a number no Indian valuer, tax officer or acquirer will recognise.

This post deliberately prints no rate. Published discount rates for Indian startups vary by sector, stage and year, and a figure quoted without a source and a date is worth less than nothing. Ask your valuer to show the build-up component by component, and where each came from. That build-up is the first thing any reviewer attacks.

Terminal value, where most of the number hides

Terminal value is the present value of everything the business does after the forecast ends. For a mature company it might be half the total. For a five-year-old startup on a five-year model it routinely accounts for most of the answer.

Sit with that. Most of the valuation is not derived from the plan you spent three weeks building. It comes from one assumption about what happens in perpetuity afterwards: usually a long-term growth rate applied to a terminal-year cash flow, or an exit multiple borrowed from comparables.

A perpetual growth rate cannot exceed the long-run growth of the economy the company operates in (which quietly caps a lot of enthusiastic models). And if terminal value is computed off an exit multiple instead, the DCF has stopped being independent of comparables. It is a comparables valuation with extra steps.

Where a DCF stops being useful

Three situations, and you can usually tell within an hour which you are in. A pre-revenue company has no cash flows to discount, so the model is a projection of a projection. A pre-product company lacks the operating data to build it from. And a company whose forecast was written for the fundraise rather than for the business has a model in which the output was chosen first.

That last case is more common than anyone admits. If the board pack plan and the DCF plan are the same file, the DCF is describing the negotiation rather than informing it.

There is also a structural limit, and it is new. For Indian income-tax purposes a DCF on unquoted equity stopped being available on 1 April 2026 (the income-tax section below has the mechanism). It can still sit behind a FEMA certificate, support a Companies Act report, or persuade an investor. Three real uses, and the tax one is no longer among them.

Comparable company analysis, multiples and precedent transactions

If a DCF argues from your own plan, comparables argue from the market's mood. The method is simple: find businesses like yours, observe what buyers pay relative to some financial metric, apply that to you, adjust for the differences.

In India the difficulty is not the arithmetic. It is that the word "like" is doing enormous work, and the data you need to defend it is largely private. So how do you build a comparable set when almost nothing genuinely compares?

Building a comparable set when the listed comps are orders of magnitude bigger

Let's be honest: listed Indian IT services and consumer internet companies are not your peers in any operational sense. They are profitable, diversified, analyst-covered and priced for liquidity. Yours is none of those. Using their multiples unadjusted is the commonest error in a founder-built deck.

Better inputs exist, though they take work. Recently funded private companies at your stage and sector, where the round was reported with both the raise and the post-money, give an ARR multiple that reflects private-market pricing. Listed small caps in adjacent sectors anchor a range. Southeast Asian companies at similar scale are sometimes closer than an Indian listed giant.

Then adjust, explicitly and in writing: illiquidity (your shares cannot be sold on a Tuesday), growth-rate differential, margin structure, and customer concentration if one client exceeds a fifth of revenue.

In practice, five to eight names with the adjustments written beside each survive scrutiny far better than twenty with none. The list is not the argument. The adjustments are.

Which multiple suits which business model

A revenue or ARR multiple suits a subscription business with predictable renewals, the default for Indian SaaS. It rewards growth and ignores profitability: fine while capital is cheap, punishing when it is not. When software multiples reset between 2022 and 2024, the businesses had not changed. What buyers would pay for the same revenue had.

A GMV multiple suits marketplaces and commerce, but flatters anyone whose take rate is thin. Think of it this way: say two companies report identical GMV, but one keeps 3 per cent of it and the other keeps 15 per cent. Those two percentages are an invented illustration chosen to make the point, not benchmarks, and nobody should quote them as market take rates.

They are not the same business, and pricing both off GMV says they are. And net revenue multiples are the honest version.

EV to EBITDA suits a business that actually has EBITDA, which excludes most companies before Series C. For services and outsourcing businesses it is the standard, because margins are stable and cash conversion real.

One rule covers all three. Never import a multiple and a metric from different places: a US ARR multiple on an Indian ARR figure compares two markets pretending to be one. And if you quote a benchmark range, quote its source and date. A multiple without a date is not a data point.

Precedent transactions, and why Indian deal data is thin

The theory is clean. Find acquisitions of companies like yours, calculate the multiple the acquirer paid, apply it. In India the practice hits a wall: most private deal terms are never disclosed.

So what can a founder actually get hold of? Less than they expect. Listed-acquirer disclosures to the stock exchanges give real consideration figures where the target is material, Registrar of Companies filings show share transfers and allotments though not the commercial logic, and Competition Commission of India orders in notifiable transactions carry deal structure.

Trade press is the tempting fourth source and the weakest. It reports deal values, but those are often estimates from unnamed sources, and treating them as data is how a comparable set becomes fiction.

A smarter strategy is to treat precedent transactions as corroboration, not a primary method at early stage. If two or three reported deals in your sector cluster around a multiple, that is a useful check. If they scatter, they were never comparable.

And remember what a deal multiple includes: a control premium, buyer-specific cost savings, sometimes an earn-out that never paid. Headline number and money changing hands are not always the same.

Which method fits which stage

Method selection is mostly a function of available data. A company with no revenue cannot be discounted; a company with four years of audited accounts should not be scored on a checklist. Stage by stage, what does that look like? Match the method to the evidence and most of the argument disappears.

Stage	What a founder has to work with	Methods that fit	What a statutory report would use
Pre-product	A team, a prototype, spend to date	Cost to duplicate, Berkus	No income-tax report on a fresh issue; the Rule 57 formula reaches a later transfer or a below-value receipt
Pre-revenue and seed	Early users, a market thesis, no cash flows	Berkus, scorecard, risk factor summation	Any methodology accepted under FEMA if a non-resident subscribes; no income-tax report on the issue itself
Series A	12 to 24 months of revenue, a growth rate	Comparable company analysis, venture capital method	Comparables or DCF for FEMA; NAV on a transfer or a below-value receipt, not on the issuing company
Series B to Series C	Unit economics, a repeatable sales motion	Comparables, precedent transactions, DCF	DCF or comparables for FEMA; NAV on a transfer or a below-value receipt, not on the issuing company
Growth and pre-IPO	Audited accounts, a visible path to profit	DCF, comparables, precedent transactions	DCF or comparables for FEMA; NAV on a transfer

Pre-revenue and seed

Here's the thing about the earliest rounds: the qualitative methods are not a compromise. They are the right tools, because what is being priced genuinely is the team, the insight and the early signal.

Which is why founders reach for SAFEs and convertible notes: instruments that take money now and set the price later, at the next priced round, usually with a discount or a valuation cap. Deferral is real and often sensible. But it postpones the number, it does not remove it. The cap is a valuation opinion wearing a different hat, and conversion crystallises a price agreed months earlier under different facts.

Worth flagging: only a fully and compulsorily convertible instrument counts as an equity instrument for Indian exchange-control purposes, and for those the price or conversion formula has to be fixed upfront at issue rather than at conversion. A startup convertible note is a separate category again, with its own minimum ticket and its own conversion window, and a US-style SAFE is not a recognised instrument for non-resident investment at all. Founders who treat a note as a way to skip valuation entirely tend to discover this at filing time.

Series A to Series C

Once there are twelve to twenty-four months of revenue, the conversation changes character. Investors stop scoring your team and start dividing your ARR into their fund's return requirement, which means comparable company analysis and the venture capital method are doing the real work, whatever the deck says.

A DCF becomes genuinely defensible somewhere around Series B for most businesses, because that is when the forecast rests on a repeatable sales motion rather than on hope. Before that, the model mainly shows you understand your own unit economics, and a sophisticated investor reads it that way.

The practical reality is that by Series B you will be running two numbers in parallel for the first time: the commercial price on the term sheet, and the statutory number a professional has to certify for the filings that follow. They are computed differently and they will not match. Expecting them to match is the source of most valuation panic at this stage.

Growth stage, secondaries and the pre-IPO round

At growth stage the methods converge, which is a relief after years of arguing. There are audited multi-year accounts, a credible forecast, real comparables, and often reported deals in the sector. DCF and comparables are both available, and we'd recommend running them side by side and reconciling the gap rather than picking the flattering one.

Two new events also arrive. Secondaries, where existing shareholders sell to incoming investors, put a price on shares that already exist and pull the transaction into a different part of the tax machinery than a fresh issue does. And exit valuation becomes a live question, because a trade sale prices control and strategic fit rather than standalone cash flow (the legal process behind an Indian startup acquisition runs on its own timetable and approvals).

The pre-IPO round is where all of this stops being theoretical, and where the cloud-kitchen company at the top of this article ran aground. Institutional buyers apply listed-market comparables to a business that has been priced on private-market comparables for six years. The two frameworks meet, and one of them gives.

The four valuation reports an Indian startup actually needs

Ask a founder how many valuations the company needs in a year and the answer is almost always one. Ask the company secretary who files the forms and it climbs to three or four, each answering to a different statute, carrying a different signature, measured as at a different date. So which of them does a given transaction actually require?

Here's the thing most founders find out late. A report commissioned for a Companies Act allotment doesn't satisfy the Reserve Bank of India, and what income-tax law now asks for is not a "report" in the sense founders mean. So work out which trigger event you're standing in before briefing anyone.

Trigger event	Statute and rule	Permitted method	Who may sign	Valuation date and validity	What happens if you get it wrong
Preferential allotment of shares	Section 62(1)(c), Companies Act, 2013, with Rule 13, Companies (Share Capital and Debentures) Rules, 2014	Valuer's professional judgement. No method prescribed.	Registered valuer on the IBBI register, securities or financial assets class	Relevant date stated in the report; no statutory validity period	Allotment price unsupported; the valuer carries personal exposure
Issue or transfer of equity instruments involving a person resident outside India	Rule 21, Foreign Exchange Management (Non-debt Instruments) Rules, 2019; RBI Master Direction paras 8.1.1, 8.2, 8.3.1	Any internationally accepted pricing methodology on an arm's length basis	Chartered Accountant, or SEBI-registered Merchant Banker, or practising Cost Accountant	As at the issue or transfer. FC-GPR within thirty days of issue	Breach of the Rule 21 pricing guidelines
Transfer of unquoted equity shares below fair market value (seller side)	Section 79, Income-tax Act, 2025, with Rule 57, Income-tax Rules, 2026	Rule 57 row 4: the net-asset-value formula, and nothing else, for equity shares. Row 5 for preference shares and convertibles: open-market price	None. It is arithmetic off an audited balance sheet	Date the share is transferred	Consideration deemed to be fair market value; the seller's gain recomputed
Receipt of shares for less than fair market value (recipient side)	Section 92(2)(m), Income-tax Act, 2025, with Rule 57	The same Rule 57 table, by class of instrument	None	Date the shares are received	Shortfall taxed in the recipient's hands
ESOP perquisite at exercise	Section 17(1)(d), Income-tax Act, 2025, with Rule 15(6)(d), Income-tax Rules, 2026	Value determined by a merchant banker	SEBI Category I merchant banker only. A Chartered Accountant may not sign	Specified date: the exercise date, or up to 180 days before it	Perquisite misvalued; the employer's withholding falls short
Sweat equity, or shares issued for consideration other than cash	Section 54, Companies Act, 2013, with Rules 8(6) and 8(7), Companies (Share Capital and Debentures) Rules, 2014 (sweat equity); Rule 13(2)(d)(xii) of the same Rules (allotment for non-cash consideration)	Fair price determined by the valuer, plus a separate valuation of the intellectual property, know-how or value addition	Registered valuer on the IBBI register	Relevant date stated in the report	Consideration unsubstantiated; the allotment open to challenge

The signature column is where the professions split: FEMA accepts a plain Chartered Accountant's certificate; income-tax law won't accept one for an ESOP. Same company, same shares, two professions. And the method column is where 2026 broke with everything published before it.

Two registered-valuer regimes, and why being one does not make you the other

India runs two separate registers of valuers, under two statutes, kept by two authorities. Most published guidance treats "registered valuer" as one profession. It isn't, and the conflation gets expensive when a report is rejected. So which register is your valuer actually on?

The first is the income-tax register. Section 514 of the Income-tax Act, 2025 requires the Principal Chief Commissioner or Chief Commissioner, or the Principal Director General or Director General, to maintain a register of valuers, who undertake to value impartially, to report in the prescribed form, to charge no more than the prescribed rate, and to avoid any asset they hold an interest in. Rule 246 of the Income-tax Rules, 2026 prescribes Form No. 169 to apply, with a non-refundable fee of Rs 10,000, and Rule 248 prescribes Form No. 170 for the report. The regime descends from the Wealth-tax Act, 1957.

Note what Form No. 169 is: the valuer's own application to join the register. A startup never files it. Nor does a startup engage section 513, which looks like a valuation mandate and is a right of representation before an income-tax authority or the Appellate Tribunal. It grants a right, not an obligation.

The second is the Companies Act register, under section 247 of the Companies Act, 2013, kept by the Insolvency and Bankruptcy Board of India. Entry runs through membership of a Registered Valuers Organisation and the IBBI valuation examination, in one of three asset classes: securities or financial assets, land and building, and plant and machinery. Startup equity sits in the first, and being on one register doesn't put you on the other.

So what does the tax register ask for? Class 6 of Rule 247 covers stocks, shares, debentures, securities and business assets including goodwill. The applicant must belong to the Institute of Chartered Accountants of India, the Institute of Cost Accountants of India or the Institute of Company Secretaries of India, or hold an MBA or postgraduate diploma in business management specialising in finance, or be a postgraduate in finance, and on top of that must clear a ten-year bar: ten years in relevant practice, with gross receipts of at least Rs 1,00,000 in three of the five preceding years, or ten years' service in audit, accounts or taxation work behind them.

Set that against the other register. Annexure IV to the Companies (Registered Valuers and Valuation) Rules, 2017 asks the same professionals, for the securities or financial assets class, for three years of experience, not ten. Same person, same equity, two registers, and the harder one to join is the one nobody writes about. And unlike the Companies Act regime, the tax register caps by statute what a valuer charges.

The events that need no report at all

Every guide tells founders when they need a valuation. Almost none says when they don't, which is the more useful half of the answer. The uncomfortable truth is that the mandatory list is a good deal shorter than the industry around it implies.

A rights issue offered pro rata to existing shareholders is the clearest case. Section 62(1)(a) of the Companies Act, 2013 runs its own procedure, and Rule 13 attaches to preferential allotment, not to a rights offer. Nobody is being favoured, so there's nothing for a valuer to defend.

On the income-tax side that list got much longer on 1 April 2026: a fresh issue by a resident company carries no valuation obligation on the issuing company at all. That's not a loophole. It's what happens when a charging provision is repealed and nothing replaces it. The subscriber's own position is a separate question, and section 92(2)(m) of the Income-tax Act, 2025 still answers it.

And if you run an LLP rather than a private limited company, none of this reaches you. An LLP can't issue securities, so it has no share fair market value.

How to brief a valuer so the report survives its purpose

Here's the mistake we see most often. A founder asks for "a valuation", receives a competent thirty-page document, and discovers four months later that it can't support the filing it was bought for.

Purpose is not a preference. The Companies (Registered Valuers and Valuation) Rules, 2017 require the report to state the purpose of the valuation and the authority appointing the valuer, so the legal purpose is written in and can't be retrofitted. A certificate scoped for FEMA pricing won't carry a Companies Act allotment.

So decide the trigger event first, then the statute, then the signatory, then commission. And give the valuer the transaction date you're working to, because that date is now fixed by statute rather than by convenience. A report drawn to the wrong one is not slightly stale. It is the wrong report.

Income-tax valuation after 1 April 2026: Rule 11UA is now Rule 57

This is the section where most published guidance is now wrong, and wrong in a way that costs founders fees they no longer need to pay. So what actually replaced the rule everyone is still citing?

The erstwhile Rule 11UA of the Income-tax Rules, 1962, cited by every Indian startup guide for a decade, ceased to apply on 1 April 2026. Its successor is Rule 57 of the Income-tax Rules, 2026, made under the Income-tax Act, 2025, and the Central Board of Direct Taxes says so in its own 1961-to-2026 mapping, which folds the erstwhile Rules 11UA, 11UAA and 11UAB into a single Rule 57.

But the successor is not a renumbered copy. It is far narrower, and the narrowing is the story.

What Rule 57 actually says about unquoted equity shares

Rule 57 sets out a table of seven rows. Row 4 covers unquoted equity shares and prescribes one manner of determination:

Fair market value of unquoted equity shares = (A + B + C + D − L) × (PV)/(PE)

In plain English:

• A is the book value of the company's assets in the balance sheet, excluding jewellery, artistic work, shares, securities and immovable property, reduced by income-tax paid (net of refund claimed) and by anything on the asset side that isn't really an asset, such as unamortised deferred expenditure.

• B is the open-market price of jewellery and artistic work, on a registered valuer's report.

• C is the fair market value of shares and securities the company holds, under the same rule.

• D is the stamp-duty value of immovable property, as adopted or assessed by the authority.

• L is the book value of balance-sheet liabilities, less six listed exclusions.

• PV is the paid-up value of the shares being valued.

• PE is total paid-up equity share capital per the balance sheet.

Notice what is missing. No cash-flow projection, no discount rate, no terminal value, no view of the future at all. The only outside opinion the formula admits is a registered valuer's report on jewellery and artistic work, which for almost every startup is zero. This is arithmetic off an audited balance sheet, and two competent people working from the same sheet should reach the same number.

Row 5 matters for most priced rounds, because it covers unquoted shares and securities other than equity: compulsorily convertible preference shares and convertibles. There the standard is the price the instrument would fetch if sold in the open market on the valuation date, and the assessee "may obtain a report from a merchant banker or an accountant".

Two words in that phrase carry weight. "May", because the report is optional. And "accountant", because the erstwhile DCF rule was merchant-banker-only.

What disappeared on 1 April 2026

So what actually went? Five things every Rule 11UA explainer still describes no longer exist for unquoted equity shares.

The DCF election is gone. Under the erstwhile Rule 11UA(2)(b) an assessee could elect to have fair market value determined by a merchant banker on the discounted free cash flow method, and Rule 57 has no equivalent. The merchant-banker certificate for unquoted equity went with it, along with the 10% safe-harbour band and the five additional methods once available for non-resident subscribers.

The fifth is quieter and bites harder. Rule 56 now defines "balance sheet", for an Indian company, as the audited balance sheet "as drawn up on the valuation date". Full stop. The erstwhile Rule 11U permitted an earlier audited balance sheet, and the 11UA proviso gave a ninety-day report window on top. Both are gone.

That is a real tightening, and the change founders are least likely to hear about. If your valuation date is 14 March and your last audited balance sheet was drawn to 31 December, no drafting fixes it.

A fresh share issue now triggers no income-tax valuation in the company's hands

Plainly: after 1 April 2026, an Indian startup issuing fresh shares at any price has no income-tax valuation obligation in its own hands.

Three things have to be true at once, and all three are. Angel tax, the erstwhile section 56(2)(viib) of the Income-tax Act, 1961, has no successor in the Income-tax Act, 2025. Section 79 of the Income-tax Act, 2025 is a transfer provision, and Rule 56 says so on its face: the valuation date for section 79 is the "date on which the capital asset, being share of a company other than a quoted share, referred to in section 79, is transferred". And section 92(2)(m) of the Income-tax Act, 2025 bites the person who receives property below fair market value, not the company that issued it.

Primary issuance sits outside the first two outright. The third needs one qualification, and it is the qualification nobody writing about the 2026 rules has made: section 92(2)(m) turns on receipt, not on transfer, and section 92(5)(f) of the Income-tax Act, 2025 lists "shares and securities" among the property it reaches.

So read the charge carefully, because it lands on one side of the table and not the other. An investor who subscribes to fresh shares for a consideration below their Rule 57 fair market value, by more than Rs 50,000, is charged on the shortfall. The company that issued them is not.

Bottom line: the whole "Rule 11UA compliance for your funding round" genre describes a regime that ended, but the exposure moved rather than disappeared. Frankly, this gets overlooked because habits outlive statutes.

Where the obligation survives: a transfer under section 79, a receipt under section 92(2)(m)

The obligation did not vanish. It moved off the company and onto individuals.

Section 79 catches a shareholder selling unquoted shares below fair market value: the founder secondary, the ESOP cashout where the employee sells rather than exercises, the early investor taking liquidity. Consideration is deemed to be the Rule 57 figure, and capital gains are computed on the deemed number rather than the agreed one, as at the date of transfer.

Section 92(2)(m) catches the other side. A recipient who acquires shares below fair market value is taxed on the shortfall, valued as at the date of receipt. Both run through the same formula, which is why a secondary at a discount to net asset value can be taxed twice over.

So a 2026 startup runs its priced round without an income-tax valuation on its own account, and the formula comes back out of the drawer the moment a founder sells a share.

Angel tax: what the sunset clause says, and what practitioners still disagree about

Angel tax ended through the Finance (No. 2) Act, 2024, which received Presidential assent on 16 August 2024. The operative words are that the clause "shall not apply on or after the 1st day of April, 2025".

That sentence is where the field splits. Indian amendments taking effect from the first day of April are conventionally read as applying from the assessment year beginning that date, which would put the last exposed transaction year at FY 2023-24. But this proviso reads as a limit on the clause's own application rather than a commencement clause, and reputable practitioners take it as clearing share issues from FY 2024-25 instead. Published guidance states both with equal confidence, which should tell you something.

So here's the practical line: if you closed a round before April 2025 and the assessment is still open, read the Memorandum to the Finance (No. 2) Bill, 2024 with your advisor rather than trusting a blog post for the year, this one included.

FEMA pricing when a non-resident is on the cap table

Here is where the certificate you actually need lives in 2026. The moment a person resident outside India appears on your cap table, exchange-control law sets a price boundary on the transaction, and that boundary has never had anything to do with income-tax rules. So which way does the boundary point, and who is allowed to certify it? For entry routes and sectoral conditions, the companion guide to foreign direct investment rules and routes covers separate ground.

The pricing guidelines sit at Rule 21 of the Foreign Exchange Management (Non-debt Instruments) Rules, 2019, restated in the RBI Master Direction on Foreign Investment in India.

Floor price, ceiling price, and which direction you are going

FEMA pricing is directional, and getting the direction wrong is the classic error.

On an issue of equity instruments to a person resident outside India, para 8.1.1 says the price "should not be less than" the benchmark. That is a floor. On a transfer from a resident to a non-resident, para 8.2 applies the same logic. But on a transfer from a non-resident to a resident, para 8.3.1 says the price "should not exceed" the benchmark. That one is a ceiling.

The policy is consistent even though the direction flips: India does not want value walking out below its worth, or money walking in above it. So ask which way the share is moving before checking any number.

For an unlisted company the benchmark itself is "valuation of equity instruments done as per any internationally accepted pricing methodology for valuation on an arm's length basis". Note what that is not. It isn't a formula, and it names no method.

Who may certify a FEMA valuation

So who can sign one? Para 8.1.1 is explicit that the valuation must be "duly certified by a Chartered Accountant or a SEBI registered Merchant Banker or a practicing Cost Accountant". Three professions, any of which will do.

That list is broader than anything income-tax law allows. A practising Cost Accountant can certify a FEMA valuation and cannot sign an ESOP perquisite valuation. A Chartered Accountant can do the first and not the second.

Two corrections are worth making here, both on points that appear in otherwise careful guidance. FEMA does not mandate discounted cash flow. DCF is what most certifiers use, but the regulation names no method, and a comparable-company approach on an arm's length basis is equally compliant. And the erstwhile Rule 11UA never governed FEMA pricing. Separate regimes, commissioned in the same week.

FC-GPR and the thirty-day clock

Form FC-GPR must be filed within thirty days from the date of issue of the equity instruments, and the valuation certificate supporting the price is part of what goes with it. Worth flagging: thirty days from issue, not from the board meeting and not from receipt of funds.

The filing chain a round runs through, from board resolution to PAS-3 to FC-GPR, is set out in the compliance checklist for startup funding rounds. What matters here is that the certificate exists before the clock runs out, which means commissioning it alongside the term sheet.

Can one report serve both FEMA and income tax?

What follows is this post's own analysis of the verified primary text rather than a proposition stated in any source.

On the old law the honest answer was no, for three reasons that still hold. The regimes test different things: FEMA asks for a forward-looking economic valuation on an arm's length basis, Rule 57 asks for a mechanical computation off an audited balance sheet, and no drafting makes those the same number. They accept different certifiers. And they run to different valuation dates.

But the 2026 position makes most of that moot. For a primary issue of shares to a non-resident, only FEMA binds the company: there is no angel-tax charge on it, and section 79 of the Income-tax Act, 2025 does not reach an issue. Section 92(2)(m) sits on the subscriber rather than the issuer, and a FEMA-compliant round is priced at or above fair value anyway, which is the direction that keeps it out of trouble. So a startup raising a foreign round in 2026 commissions one certificate, not two.

The practical upshot is that the old "two reports" problem was an artefact of angel tax. Where you still need two is a secondary sale to a non-resident, because a transfer engages both the FEMA floor and section 79.

The Companies Act layer: section 247 and the registered valuer

Company law is the layer founders most often forget: it did not change in 2026, and their investors' counsel handles it quietly. It still governs how a priced allotment gets approved, and it is the only one of the three that puts the valuer personally in the dock, with a criminal limb and a statutory damages remedy attached.

So when does a startup need a section 247 valuer? Whenever the Companies Act requires a valuation of property, stocks, shares, debentures, securities, goodwill, other assets, net worth or liabilities.

Who may be appointed, and by whom

So who signs, and who hires them? Under section 247 of the Companies Act, 2013, the valuer must be qualified, experienced and registered as prescribed, and appointed by the audit committee, or in its absence by the Board of Directors. Most early-stage private companies fall into the second limb, so this is a board resolution.

The valuer owes statutory duties: an impartial, true and fair valuation, due diligence in performing it, and compliance with the prescribed rules. Section 247(2) also bars a valuer from valuing any asset in which they hold a direct or indirect interest, or become so interested at any time during or after the valuation.

That bar is the one to watch on a small cap table. A valuer who advises the company on other matters is not automatically conflicted, but the question has to be answered on the record before the appointment, not after the report lands.

Preferential allotment under section 62(1)(c) and Rule 13

This is the report a private priced round actually runs on, so the rule is worth reading closely. A preferential allotment under section 62(1)(c) of the Companies Act, 2013, read with Rule 13 of the Companies (Share Capital and Debentures) Rules, 2014, requires the price to be supported by a registered valuer's report. The rule carries exactly one carve-out, and it is not the one founders hope for: the price of shares issued on a preferential basis "by a listed company shall not be required to be determined by the valuation report of a registered valuer", because SEBI's ICDR pricing applies to them instead.

There is no equivalent relief for a private company. Rule 13(2)(d)(iv) goes the other way, requiring the explanatory statement sent to members to disclose the "basis on which the price has been arrived at along with report of the registered valuer".

For a private company raising a priced round, that is the report the filing chain runs on. It's also the one most likely to be commissioned late, because founders assume the investor's diligence valuation covers it. It doesn't.

What a registered valuer's report has to contain

The Companies (Registered Valuers and Valuation) Rules, 2017 set out the content list at Rule 8(3), and it runs to twelve items, from (a) to (l): background information of the asset being valued; the purpose of the valuation and the appointing authority; the identity of the valuer and of any other experts involved; disclosure of valuer interest or conflict, if any; the date of appointment, the valuation date and the date of the report; inspections and investigations undertaken; the nature and sources of the information relied upon; the procedures adopted and the valuation standards followed; restrictions on use of the report, if any; the major factors taken into account; the conclusion; and, last, caveats, limitations and disclaimers, but only "to the extent they explain or elucidate the limitations faced by valuer, which shall not be for the purpose of limiting his responsibility for the valuation report".

Read that as a buyer rather than a compliance officer and it tells you how to evaluate a draft. Three dates, a named purpose, a conflict disclosure, a statement of what the valuer relied on. If the report you are paying for is thin on any of those, it is thin in the wrong places.

What happens if the valuer is not registered

The answer is harsher than most founders expect. A defective appointment doesn't produce a weaker report: it produces a document that isn't a section 247 valuation at all, so the allotment it supports rests on nothing.

The section attaches consequences to the valuer personally too. Section 247(3) sets a fine of not less than Rs 25,000, extending to Rs 1,00,000. Where the contravention is committed "with the intention to defraud the company or its members", the exposure becomes criminal: imprisonment for a term which may extend to one year, and a fine of not less than Rs 1,00,000 extending to Rs 5,00,000. And on conviction, section 247(4) requires the valuer to refund the remuneration received to the company and to pay damages to the company or to any other person for loss arising out of incorrect or misleading statements of particulars in the report.

For a founder the takeaway is duller and more useful: ask for the registration number and asset class before you engage, and keep a copy in the round file. It is the first thing an acquirer's counsel checks three years later.

ESOP valuation: the one place a merchant banker is now mandatory

Now, here is where it gets interesting. The merchant banker used to be the funding round's valuation professional, signing the DCF certificate that supported the share premium. After 1 April 2026 the merchant banker is out of the round entirely, and mandatory in the one place most founders treat as an afterthought. So where did the profession go?

What follows covers the fair market value determination at exercise only. For pool sizing, vesting, cliffs and tax treatment, the guide to ESOP structuring for Indian startups handles the structure.

Rule 15 and the fair market value at exercise

Start with the charge itself. The charging provision is section 17(1)(d) of the Income-tax Act, 2025, and the perquisite is the fair market value of the share on the date the option is exercised, less what the employee actually paid for it.

Rule 15 of the Income-tax Rules, 2026, successor to the erstwhile Rule 3 of the 1962 Rules, sets out how that value is found. Rule 15(6)(d) is the operative limb: where, on the date of exercise, the share is not listed on a recognised stock exchange, the fair market value "shall be such value of the share in the company as determined by a merchant banker on the specified date". Rule 15(7) applies the same requirement to specified securities that are not equity shares.

"Merchant banker" is not loose language here. Rule 15(8)(g) defines it, for the purposes of Rule 15 itself, as a Category I merchant banker registered with the Securities and Exchange Board of India, the same formula Rule 56(b) uses for Rules 56 and 57. A Chartered Accountant may not sign this valuation, and neither may a Cost Accountant.

Set that against FEMA, which accepts all three. Same company, same equity, two statutes, and only one will take your auditor's signature.

The 180-day window, and how to use it

Rule 15(8)(l) defines the "specified date" as the date of exercise, or any earlier date not more than one hundred and eighty days before it.

Read that as an operational instrument rather than a definition, because it is one. A single merchant-banker valuation can support every exercise in the following six months, which makes the cost of ESOP valuation a function of how you schedule exercises, not how many employees exercise.

So run exercise windows in batches. A company that opens two windows a year, each priced off one valuation obtained shortly before it opens, pays for two reports. One that lets employees exercise at will may commission four or five, for no extra compliance benefit.

Fair warning: the 180 days run backwards from exercise, not forwards from the report. A valuation dated 1 January covers exercises to the end of June, and an exercise on 5 July needs a fresh one.

Why the ESOP price sits below the round price, and how to defend the gap

Employees who read the funding announcement ask a fair question: if the company just raised at Rs 400 a share, why does my option price off a lower number?

Because they are different instruments. Investors in a priced round almost always buy compulsorily convertible preference shares carrying a liquidation preference, anti-dilution protection and usually consent rights. An ESOP delivers ordinary equity, which sits behind all of that in a distribution and carries none of the protections. The preference share is worth more because it is worth more, not because anyone is discounting the employee.

What makes the gap defensible is the merchant banker's report saying so in terms, and a smarter approach is to ask for that reasoning rather than accept a one-page certificate with a number on it. And explain the logic at grant, not at exercise, when the conversation has become about tax.

What the tax tribunals have held about DCF, and whether it still matters

Founders ask a version of the same question every time a valuation goes into a return. Can the assessing officer simply refuse to accept it? For a decade that answer came from the Income Tax Appellate Tribunal rather than the rule book, and it was largely no.

Here's the thing. That body of orders is worth reading, and it's also worth knowing which parts survived 1 April 2026.

The principle: the choice of method is the assessee's

The cleanest recent statement comes from the Income Tax Appellate Tribunal, Delhi Bench "G", in ITA No. 1746/Del/2025, an order pronounced on 14 November 2025 for assessment year 2022-23. The tribunal upheld the deletion of an addition of Rs 52,68,21,209. Worth flagging: the addition was made under the erstwhile section 50CA, the transfer provision whose successor is section 79 of the Income-tax Act, 2025, not under angel tax.

The reasoning matters more than the number. The finding the tribunal affirmed, reproduced in full in the order, is that the rule "gives the option to assessee to follow either NAV method or DCF method", and goes further: "there is no bar on the assessee to use different methods in same AY on sale of same company's unlisted shares". Once a value had been determined by one of the two permitted methods, that value was deemed to be fair market value, and the "AO cannot question the same" unless he "shows mistake in the method adopted by the assessee". The Revenue's appeal was dismissed, and the full order sits on Indian Kanoon.

In practice, a shareholder selling two tranches of the same unlisted company in one year could price one on net asset value and the other on discounted cash flow, and the officer could not harmonise them.

Projections against actuals, and the hindsight problem

Sitting behind that order is a longer line of authority, and its effect is easy to state. A tax officer may not discard a valid valuation and substitute a number of his own without showing what was wrong with the method, and may not rewrite a forecast by replacing the projections with what actually happened. The 2021 decision of the High Court of Delhi in ITA 1007/2019, reported at (2021) 433 ITR 82, is the authority usually cited: challenging a valuation because performance later diverged from the forecast "lacks material foundation and is irrational since the valuation is intrinsically based on projections which can be affected by various factors", and "valuation is not an exact science, and therefore cannot be done with arithmetic precision". The recent restatement is ITA No. 6291/Mum/2025, a Mumbai "C" bench order pronounced on 15 May 2026 for assessment year 2017-18, which vacated a share-premium addition of Rs 36,54,46,130 and held that the officer "ought not to have disturbed the method adopted by assessee or cannot adopt the figures by replacing the projections with actual financials".

Think of it this way. A forecast made in March 2022 was either reasonable on what was knowable then or it wasn't, and a market turning in 2023 says nothing about that. Assessing a projection against the outcome is marking the homework with the answer sheet.

A question that comes up constantly on Quora and in founder groups is whether this makes a DCF bulletproof. It doesn't, and reading it that way is a serious misreading of what those orders decided. They address the officer's power to substitute a method. They are not a licence for assumptions nobody can support.

Does the case law survive Rule 57?

Let's be honest: this is the post's own analysis, not settled law. But the question is unavoidable, and nobody in the field appears to have asked it.

The entire line of authority rests on one structural premise: that the erstwhile Rule 11UA(2) gave the assessee an election between net asset value and discounted cash flow. The Delhi tribunal order says exactly that, in terms. Rule 57 of the Income-tax Rules, 2026 contains no election for unquoted equity shares: there is one method, and it is arithmetic off an audited balance sheet. Remove the election and the ratio has nothing to bite on.

Three consequences follow, in descending order of confidence. First: for transactions in unquoted equity shares on or after 1 April 2026, the DCF-rejection case law is largely spent, because there is no projection left for an officer to second-guess. The disputes that replace it will be about balance-sheet inputs instead: what belongs on the asset side, and which liabilities are excluded.

Second: the case law stays live for legacy and pending assessments for years yet. Pre-April-2026 transactions are still governed by the old rule, tribunal backlogs are what they are, and the Delhi tribunal order itself was decided in November 2025 on assessment year 2022-23.

Third, and this one is an open question rather than an answer. Row 5 of Rule 57, covering unquoted shares and securities other than equity, uses an open-market standard with an optional expert report, which sits closer to the old posture than row 4 does. Whether a court carries the Rule 11UA reasoning across is genuinely unlitigated. This post won't predict it, and neither should anyone selling you a report.

What a valuation report costs, and how long it takes

Ask the internet what a startup valuation costs in India and the answers differ by more than a hundredfold. Why would one service vary that much? It wouldn't, and it isn't one service: six different products are sold under a single name, and almost nobody separates them.

Count the six. A Companies Act registered valuer's report, an income-tax registered valuer's report, a FEMA pricing certificate, a merchant banker's ESOP valuation, a deal-support model built for a negotiation, and full transaction advisory. There is also a source almost nobody cites for any of them, and it's primary law rather than a price page.

The statutory fee cap nobody cites

Rule 248 of the Income-tax Rules, 2026 caps what a registered valuer may charge. The fee "shall not exceed" 0.5 per cent of the first Rs 5,00,000 of the asset as valued, 0.2 per cent of the next Rs 10,00,000, 0.1 per cent of the next Rs 40,00,000, and 0.05 per cent of the balance. Where that produces less than Rs 5,000, the valuer may charge Rs 5,000.

Two details are easy to miss. Where two or more assets are valued for the same assessee, they are deemed to be a single asset for the fee calculation, so a valuer cannot slice one engagement into several to climb back up the scale. And the report goes in Form No. 170.

Run the scale yourself and the numbers are modest. On an asset valued at Rs 1 crore the statutory maximum works out to Rs 10,750, and on Rs 10 crore it is Rs 55,750. Both are worked examples, arithmetic off the statutory scale in Rule 248 rather than prices anyone has quoted or charged.

Now the scope limit, because without it this section misleads. Frankly, the cap goes uncited because the register it belongs to goes unexplained.

It governs the income-tax registered valuer registered under section 514 of the Income-tax Act, 2025. It does not price a Companies Act registered valuer's report under section 247, a merchant banker's ESOP valuation, or a chartered accountant's FEMA pricing certificate. And those three are unregulated as to fee, which is why the ranges below are so wide.

What the market actually charges, and why the published figures disagree

Let's be honest: there is no market price for any of those six, only five published sources whose claims contradict each other. Here is the whole field, each row attributed and dated.

Source and date	Registered valuer	Merchant banker	Turnaround quoted
CA firm with an IBBI registered-valuer practice, Feb 2026	Rs 15,000 to Rs 5,00,000 and above, by approach	Not quoted	1 to 6 weeks
EzyLegal, Mar 2025	From Rs 25,000	From Rs 65,000	4-5 business days; about 8 for a merchant banker
CalcGuru, Aug 2026	Rs 25,000 to Rs 75,000	Rs 75,000 to Rs 1,50,000	Not quoted
EquityList, Jul 2026	Not quoted	Not quoted	5-10 working days; 7-15 for a merchant banker
Dealplexus, Apr 2026	Rs 2,00,000 to Rs 25,00,000	Not quoted	Not quoted

The spread runs from Rs 15,000 to Rs 25,00,000, and picking a single figure out of that range and calling it the answer would be inventing data. The top looks like enterprise transaction advisory rather than a Series A share valuation; the bottom looks like a compliance filing with a formula in it.

What drives the number, and the turnaround to plan for

So what actually drives the quote? Five things move the price, and stage is not really one of them. What matters is the asset class the valuer is registered for, whether a model must be built from scratch or the answer is a formula off an audited balance sheet, which filing the report must survive, urgency, and how many reports the event actually requires. A resident-only preferential allotment needs one registered valuer's report; a round with a non-resident needs a pricing certificate too.

In practice, urgency is the most expensive of the five and the most avoidable. A report commissioned against next week's board date carries a premium (sometimes a steep one) that one commissioned six weeks out does not.

On timing, the published figures again disagree. One source gives 4 to 5 business days for a registered valuer and about 8 for a merchant banker; another gives 5 to 10 working days and 7 to 15 working days for the same two. Bottom line: treat the longer end as your planning assumption. The common pitfall is budgeting the valuer's turnaround and forgetting the audited balance sheet the income-tax formula now needs as at the valuation date, which is often what actually holds the round up.

When the valuation number and the round price disagree

Every guide on this subject says the same reassuring thing: the commercial valuation and the statutory fair market value are different numbers, so don't be alarmed when they diverge. True, and useless. So what does a founder actually do when the two numbers pull apart?

Now, here's where it gets interesting. The gap has a direction, and the direction decides who it bites. A round priced above fair value and one priced below are not two versions of the same problem; they land on different parties, under different statutes.

And since 1 April 2026 the dangerous direction has flipped. Most published guidance still warns about the old one.

The round price is above fair value

So is a high round price still a problem? Barely, and that is genuinely new. For a decade this was the angel-tax event: a company issuing shares to a resident above fair market value was taxed on the excess as income, which is why "Rule 11UA compliance" became shorthand for valuation in a funding round.

That charge has no successor in the Income-tax Act, 2025. For a resident round priced above fair value in 2026, the position at company level is now essentially nothing, because a fresh issue sits outside both section 79 (a transfer provision) and section 92(2)(m) (which bites a recipient).

For a non-resident round, pricing above fair value is equally fine, and for a structural reason. The FEMA pricing guidelines set a floor for a non-resident's entry, not a ceiling. A foreign investor paying more than the certified fair value has paid more than the regulator required, which breaches nothing.

What most people miss: one consequence does travel with a high round price. The ESOP exercise price gets harder to defend, because a fresh determination at exercise under Rule 15 drifts upward alongside the round, and the perquisite charge lands on the employee.

The round price is below fair value, which is now the dangerous direction

Here's the inversion. Pricing low used to be the safe, conservative choice. In 2026 it is the one that creates exposure (in three separate places at once).

On the exchange-control side, an issue to a person resident outside India below the certified fair value breaches the pricing guidelines restated at paragraph 8.1.1 of the Reserve Bank of India's Master Direction on Foreign Investment. The floor is not advisory. A round that clears under the certificate is a compounding matter, not a drafting problem.

On the tax side, two provisions catch it from opposite ends. Section 92(2)(m) of the Income-tax Act, 2025 taxes a recipient who receives property for inadequate consideration, which is the incoming shareholder rather than the company. Section 79 catches a seller of unquoted shares in a secondary, deeming fair market value to be the consideration. So in a secondary at a soft price, the seller is taxed on money not received and the buyer on a benefit not taken in cash.

What experienced practitioners do here is unglamorous and effective: get the valuation before the price is agreed, not after. A term sheet negotiated in ignorance of the floor may have to be repapered, and that costs goodwill founders at this stage cannot spare.

The valuer's number comes in under the term sheet

So the report lands below the price the investor has already agreed. What now?

First, be clear about which way that cuts. A valuation below the round price is the benign direction, and for a resident round in 2026 it usually needs no action. The problem case is the reverse: a valuation above the agreed price on an issue to a non-resident, which puts the round under the FEMA floor.

Whether the report can be re-scoped is the next question, and the honest answer is that re-scoping is legitimate and re-pricing is not. A valuer can properly be asked to correct a factual input, to value as at a different date, or to value a different instrument, and any of those may change the number.

But a valuer asked to reach a predetermined figure is being asked to breach the duty section 247(2) of the Companies Act, 2013 puts on him to make an impartial, true and fair valuation. The statutory penalty for that falls on the valuer, not on the company. And the company's exposure is different and quieter: it is left holding a report that will not survive the first person who reads it properly.

But what a founder can legitimately do is narrower, and real. Brief the valuer properly, supply what the method actually needs, choose the valuation date deliberately rather than by accident, and if the instrument is a convertible rather than ordinary equity, say so, because a different row of Rule 57 applies.

The report goes stale before the board meets

The old comfort here was a 90-day window, and it is gone. Rule 56 of the Income-tax Rules, 2026 requires the balance sheet to be one "as drawn up on the valuation date", with no relaxation for an earlier audited set.

In practice, that's a scheduling problem more than a legal one. If your accounts are three months old, the income-tax number is not a little stale, it is computed off the wrong input. And a material event between the report date and the board meeting (a large impairment, a fresh round at a different price, losing the customer that was 40 per cent of revenue) can undermine a report only days old.

If your holding company is in Delaware: 409A alongside Rule 57

Flipped structures are common enough now that this comes up in most cap-table conversations, usually as a hopeful question. There's a 409A on file, so doesn't that cover India too?

The short answer is no, and the reason is structural rather than technical.

Two entities, two standards, two numbers

Here's why. A section 409A valuation is a determination of the fair market value of the common stock of the US company, prepared under US tax rules for the purpose of setting option strike prices there. An Indian determination under Rule 57 of the Income-tax Rules, 2026 or under the FEMA pricing guidelines is a determination about the Indian company's own shares, under Indian tests, for Indian filings.

Different subject, different standard, different purpose. The two can legitimately produce numbers that look inconsistent to a board, and usually do. A US valuation of common stock will typically sit below the preferred price in the last round; an Indian income-tax number for unquoted equity is now a formula off an audited balance sheet and may sit below both.

Bottom line: neither report is evidence about the other. An Indian regulator will not accept a US valuation because it exists, and a US auditor will not take comfort from an Indian registered valuer's report. Trying to make one document do both jobs produces a document that does neither.

There is one US number worth knowing, and it is in the regulation rather than on a vendor page. The independent-appraisal presumption in the section 409A regulations runs off a valuation "as of a date that is no more than 12 months before the relevant transaction", and the presumption is rebuttable where the value "was calculated with respect to a date that is more than 12 months earlier than the date for which the valuation is being used". That is a US rule about a US determination. It says nothing about how long an Indian report lasts.

What a flip does to the Indian entity's valuation obligations

Very little, and that surprises people. Once the Indian company becomes a subsidiary of a foreign holding company, it is a company with a non-resident shareholder, which means the FEMA pricing guidelines now apply to its share issues in a way they may not have before the flip.

The Companies Act triggers are untouched. A preferential allotment by the Indian company still runs on section 62(1)(c) and still needs a registered valuer's report (whoever owns the shares). If the Indian entity grants its own options to its own employees, the exercise-date determination under Rule 15 still needs a merchant banker, regardless of what the parent does in Delaware.

A common question from founders mid-flip is whether the Indian entity can stop doing local valuations once options are granted at the parent level. Sometimes yes, if it genuinely issues nothing and grants nothing. But the moment it issues a single share (to the parent, for funding, or to a local angel who stayed behind), the Indian chain restarts. Treat the Indian entity as a company in its own right, because the statutes do.

Common valuation mistakes, and what a down round does to them

The mistakes founders make in this area are not exotic. They are four recurring assumptions, and each one is cheap to fix before the fact and expensive to fix afterwards. So which four?

The 2026 funding market makes all of them worse, because a company whose last priced round is no longer supportable has to reconcile a falling commercial number with filings made on the old one.

Treating the FEMA number and the tax number as interchangeable

Here's the confusion that causes the most rework. The two test different things and they are certified by different people: FEMA asks for a fair value arrived at on any internationally accepted pricing methodology on an arm's length basis, certified by a chartered accountant, a SEBI-registered merchant banker or a practising cost accountant. Income-tax law, for unquoted equity shares, now asks for a formula off an audited balance sheet, with no certifier at all.

But the mistake is reaching for the FEMA certificate when the Registrar wants a Companies Act report, or offering the income-tax formula to a banker who asked for a pricing certificate. Each report is scoped to a purpose, and one scoped to the wrong purpose is not a weaker report, it is the wrong document.

Running on a report that a material event has overtaken

Reports do not carry a warning label. A registered valuer's report prepared for a round in January is still a well-drafted document in July, and a founder reusing it for a September allotment usually has no sense that anything has changed.

Fair warning: it surfaces later, and usually at the worst moment. A stale or mis-scoped valuation report is one of the standard findings in a diligence exercise run by an incoming investor or acquirer, and it tends to come up at the point where a company has the least room to argue about it. The fix costs nothing: keep a one-line register of every valuation report, its purpose, its valuation date and the filing it supported.

Modelling a round without the option pool

Here's how this one plays out. A founder computes dilution on the investment against the post-money valuation, gets a clean number, and then discovers the pool expansion the term sheet required sits inside the pre-money. The dilution lands on the existing shareholders, not on the incoming investor.

That is arithmetic, not law, and it belongs in the model before the term sheet is signed. Run the pool at the size the board will actually need for the next 18 months, and run it in the pre-money if that is what the sheet says.

What a down round does to your last registered valuer report

A down round does not invalidate an earlier report, but it does something more awkward: it creates a documented, contemporaneous opinion that the company is worth less than the number in a filing you have already made. Anti-dilution clauses then recalculate earlier investors' conversion ratios, the cap table moves, and the arithmetic behind the last report no longer describes the company.

The point most founders have not thought through is what a down round does to a secondary. A founder selling unquoted shares at the new, lower price is squarely inside section 79 of the Income-tax Act, 2025, which deems fair market value to be the consideration for computing the gain. If the formula value off the audited balance sheet sits above the price the secondary cleared at (entirely possible for a company with cash on the balance sheet and a written-down commercial value), the seller is taxed on a gain that is partly notional.

So is it better to wait? Frankly, that gets overlooked in favour of arguing about the headline number. Run the section 79 arithmetic before agreeing the secondary price, because the tax cost of a founder liquidity event in a soft market is often larger than the discount being negotiated.

Frequently asked questions

What are the startup valuation methods used in India?

Two families. Qualitative methods (Berkus, scorecard, risk factor summation) price pre-revenue companies on team, product and market. Quantitative methods (discounted cash flow, comparables, precedent transactions, venture capital method, net asset value) price on cash flows and multiples.

Is a startup valuation legally required in India?

Not always. It depends on the event. A preferential allotment under the Companies Act, an issue or transfer involving a non-resident under FEMA, and an ESOP exercise each require a report. A rights issue offered pro rata to existing shareholders generally does not.

What is the difference between pre-money and post-money valuation?

Pre-money is the agreed value of the company before the new money lands. Post-money is pre-money plus the investment. The investment divided by the post-money gives the investor's percentage. An option pool expanded inside the pre-money dilutes existing holders.

Rule 11UA vs Rule 57: what changed on 1 April 2026?

Rule 11UA of the 1962 Rules became Rule 57 of the Income-tax Rules, 2026. For unquoted equity shares, the discounted cash flow election, the merchant-banker certificate, the 10 per cent safe harbour and the five non-resident methods are gone. One formula remains.

Do startups still need a valuation report after angel tax was abolished?

For some events, yes. Angel tax has no successor, so a fresh issue carries no income-tax valuation duty on the issuing company. But FEMA, the Companies Act and ESOP exercises each still require their own report, and a transfer or a receipt below fair value is still caught.

Do I need a valuation report if I am issuing shares to existing investors?

Usually not, where it is a rights issue offered to all existing shareholders in proportion to their holdings. A preferential allotment to selected investors is different: that runs on section 62(1)(c) and needs a registered valuer's report.

Can founders sell unquoted shares below fair market value in 2026?

They can price where they like commercially, but the tax follows. Section 79 of the Income-tax Act, 2025 deems fair market value to be the sale consideration for the seller, and section 92(2)(m) can tax the buyer on the shortfall.

Is a valuation report mandatory under FEMA for foreign investment?

Yes, where a non-resident is subscribing or transferring. The RBI Master Direction on Foreign Investment requires an unlisted company's equity instruments to be valued on an internationally accepted pricing methodology on an arm's length basis, and duly certified.

What is the minimum price for issuing shares to a foreign investor in an Indian startup?

There is no fixed rupee minimum. The price must not fall below the fair value certified for the unlisted company using an internationally accepted pricing methodology on an arm's length basis. FEMA sets a floor for entry, not a ceiling.

When is an ESOP valuation report required under Indian law?

At exercise. Rule 15(6)(d) of the Income-tax Rules, 2026 requires the fair market value of an unlisted company's share on the exercise date to be determined by a merchant banker, for the perquisite charge under section 17(1)(d).

Who can certify a startup valuation in India?

It depends on the statute. FEMA accepts a chartered accountant, a SEBI-registered merchant banker or a practising cost accountant. The Companies Act needs a registered valuer under section 247. An ESOP perquisite valuation needs a SEBI Category I merchant banker.

Who qualifies as a registered valuer in India for startup equity?

Two separate registers exist. The Companies Act valuer is registered with IBBI after RVO membership and an examination. The income-tax valuer is registered under section 514 of the Income-tax Act, 2025 and reports in Form No. 170. Neither qualifies you for the other.

What is the difference between a registered valuer and a merchant banker?

A registered valuer is registered under a statute (with IBBI for the Companies Act, or under section 514 for income tax) and signs Companies Act reports. A merchant banker is SEBI-registered and is the only signatory Indian tax law accepts for an ESOP valuation.

Can a chartered accountant still sign a startup valuation report?

Yes, for some purposes. A chartered accountant may certify a FEMA pricing valuation and may be an income-tax registered valuer. A chartered accountant may not sign an ESOP perquisite valuation for an unlisted company, which is merchant-banker territory.

Registered valuer, merchant banker or chartered accountant: who signs what?

FEMA: chartered accountant, SEBI-registered merchant banker or practising cost accountant. Companies Act preferential allotment: registered valuer under section 247. ESOP exercise: SEBI Category I merchant banker only. Unquoted equity under Rule 57: a formula.

How much does a startup valuation cost in India?

Published figures disagree by more than a hundredfold, running from about Rs 15,000 to Rs 25,00,000 depending on the report and the provider. For an income-tax registered valuer, Rule 248 caps the fee on a sliding scale, with a floor of Rs 5,000.

How long does it take to get a valuation report?

Published figures range from 4 to 5 business days up to 5 to 10 working days for a registered valuer, and roughly 7 to 15 working days for a merchant banker. Sources differ, and a model built from scratch takes longer than a formula.

How long is a valuation report valid?

There is no single answer. Income-tax law now pegs to a balance sheet drawn up on the valuation date. An ESOP valuation may be dated up to 180 days before exercise. A material event can make any report unusable regardless of its age.

What happens if the valuation and the investor's round price do not match?

The direction matters. Pricing above fair value is largely harmless for a resident round in 2026. Pricing below fair value breaches the FEMA floor on an issue to a non-resident, and can trigger section 92(2)(m) or section 79 on the tax side.

Can the income-tax officer reject my startup's valuation?

For pre-April-2026 transactions, tribunal orders held that the choice between the permitted methods was the taxpayer's and the officer could not substitute his own. Rule 57 now prescribes one formula for unquoted equity, so disputes shift to the inputs.

References

Statutes and rules

1. Income-tax Act, 2025: section 79 (transfer of unquoted shares below fair market value, successor to section 50CA of the 1961 Act); section 92(2)(m) (property received for inadequate consideration); section 17(1)(d) (ESOP perquisite). incometaxindia.gov.in

2. Income-tax Act, 2025, section 513 and section 514: registered valuers, registration and fee framework.

3. Income-tax Rules, 2026, Rule 56: definitions, the balance sheet "as drawn up on the valuation date", merchant banker as SEBI Category I, and the valuation-date table. Rule 56

4. Income-tax Rules, 2026, Rule 57, row 4 (unquoted equity shares) and row 5 (unquoted shares and securities other than equity shares). Rule 57

5. Income-tax Rules, 2026, Rule 15(6)(d), 15(7), 15(8)(g) and 15(8)(l): merchant-banker determination on the specified date, not more than 180 days before exercise. Rule 15

6. Income-tax Rules, 2026, Rule 246 (application in Form No. 169), Rule 247 (asset classes) and Rule 248 (statutory fee scale, and the report in Form No. 170).

7. Companies Act, 2013, section 247: appointment of the registered valuer, the duty to make an impartial, true and fair valuation, and the consequences of contravention.

8. Companies Act, 2013, section 62(1)(c) read with Rule 13 of the Companies (Share Capital and Debentures) Rules, 2014: preferential allotment priced on a registered valuer's report, and, at Rule 13(2)(d)(xii), allotment for consideration other than cash. Sweat equity runs separately, on section 54 of the Act with Rules 8(6) and 8(7) of the same Rules.

9. Companies (Registered Valuers and Valuation) Rules, 2017, Rule 8(3): required contents of a registered valuer's report. Annexure IV: the securities or financial assets class and its three-year experience requirement.

10. Foreign Exchange Management (Non-debt Instruments) Rules, 2019, Rule 21, pricing guidelines. Cited to the rule itself and deliberately left unlinked: no stable primary URL for the consolidated rule text was found at the date of this post, and the same pricing guidelines are restated verbatim in the Reserve Bank of India Master Direction linked under Regulator guidance below.

    
    

Regulator guidance

1. Reserve Bank of India, Master Direction on Foreign Investment in India, paragraphs 8.1.1, 8.2 and 8.3.1: floor and ceiling prices, and certification by a chartered accountant, SEBI-registered merchant banker or practising cost accountant. rbi.org.in

2. Central Board of Direct Taxes, section-and-rule mapping from the Income-tax Act, 1961 and Rules, 1962 to the Income-tax Act, 2025 and Rules, 2026. The mapping folds the erstwhile Rules 11UA, 11UAA and 11UAB into a single Rule 57, and maps section 50CA to section 79. incometaxindia.gov.in

3. Insolvency and Bankruptcy Board of India, register of registered valuers: the Companies Act register, its three asset classes and the valuation examination.

4. United States Treasury, section 409A regulations, 26 CFR 1.409A-1: the independent-appraisal presumption and its 12-month limit.

   
   

Case law

1. Income Tax Appellate Tribunal, Delhi Bench "G", ITA No. 1746/Del/2025, order pronounced 14 November 2025, assessment year 2022-23. Deletion of an addition of Rs 52,68,21,209 made under the erstwhile section 50CA upheld; the Revenue's appeal dismissed. indiankanoon.org

2. Income Tax Appellate Tribunal, Mumbai Bench "C", ITA No. 6291/Mum/2025, order pronounced 15 May 2026, assessment year 2017-18. Share-premium addition of Rs 36,54,46,130 under section 56(2)(viib) vacated; an assessing officer may not replace the projections in a discounted-cash-flow valuation with actual financials. indiankanoon.org

3. High Court of Delhi, ITA 1007/2019, judgment pronounced 1 March 2021, reported at (2021) 433 ITR 82. Rejecting a valuation because performance later diverged from the forecast "lacks material foundation and is irrational"; valuation "is not an exact science". indiankanoon.org

   
   

Reported data

1. CalcGuru, reviewed August 2026: registered valuer and merchant banker fee bands.

2. EquityList, 22 July 2026: registered valuer and merchant banker turnaround times.

3. Inc42, 8 June 2026: a cloud-kitchen company's deferred public listing.

4. Outlook Money, 8 June 2026: corroboration of the same deferral, and the wider pattern.

5. Dealplexus, 7 April 2026: formal valuation report cost range.

6. Chartered accountancy firm with an IBBI registered-valuer practice, 27 February 2026: cost and turnaround by valuation approach.

7. EzyLegal, updated 6 March 2025: registered valuer and merchant banker cost and turnaround compared.

8. TechCrunch, 7 June 2024: the same holding written down to zero.

9. TechCrunch, 11 January 2024: a US asset manager's 95 per cent markdown of an Indian edtech holding.

10. TechCrunch, 10 May 2023: a 35 per cent markdown of an Indian mobility company.

11. TechCrunch, 8 May 2023: a further markdown of an Indian food-delivery company to USD 5.5 billion.

12. TechCrunch, 31 March 2023: US holders' unilateral markdowns of Indian edtech and food-delivery companies.

    
    

Income-tax Department pages on incometaxindia.gov.in, and mca.gov.in, return 403 to automated link checkers but resolve in a browser, so a dead-link report on those URLs is a false positive. The cost figures are published commercial claims, not official statistics.

This article is for educational and general business information purposes only and does not constitute professional legal, financial or tax advice. It describes the Indian position on startup valuation as it stood on the date at the top of this page.

The Income-tax Rules, 2026 are recent, and rule numbering should be confirmed against the current consolidated text before anything here is acted on. Anyone pricing a round, signing a filing, granting options or selling shares should consult a qualified chartered accountant, company secretary, registered valuer or lawyer on their own facts.