Fintech valuation in India turns on a question that the label “fintech” hides: what does the company actually keep from each rupee it handles, and how much of that income depends on credit, float, a licence or a single partner bank? A payments processor, a digital lender, a wealth platform and a neobank can carry the same label and still deserve very different multiples. This guide sets out the eight drivers we test before accepting a revenue or earnings multiple for an Indian fintech. It tests the benchmarks against the 2026 public multiples, a disclosed M&A index and reported Razorpay targets and Pine Labs IPO figures, and it sets out the RBI, FEMA and data protection rules that change the price.
Key takeaways
- Start with the business model. Payments, lending and wealth platforms trade on different bases. The public payments median was 2.2x EV to 2026E revenue and 7.3x EV to 2026E EBITDA at 30 June 2026, as reported by Windsor Drake from Houlihan Lokey’s data.
- Averages mislead. The mixed-sample fintech average of 14.5x is pulled up by outliers. The median of 7.6x across 416 companies is the more useful reference, and it still says nothing about a company’s own take rate, credit cost or licence status.
- Lenders are valued on book, not on EBITDA. For a lender, interest is the cost of its raw material, so price-to-book and return on equity on an ECL-adjusted book are the anchors.
- Reported Razorpay IPO targets show why the arithmetic and value basis must be checked. Its target of ₹50,000 crore to ₹60,000 crore on FY25 revenue of ₹3,783 crore works out to about 13x to 16x. The cited media report states 1.3x to 1.6x, which does not reconcile with those rupee figures. A target equity value divided by revenue is not an enterprise-value multiple.
- Indian rules are value drivers, not compliance footnotes. RBI authorisation, net worth, escrow, the digital lending directions and the DPDP timelines all change what a buyer can earn. Rule 57 NAV for tax is a separate number from the commercial price.
For a transaction-specific scope, explore our business valuation services. For adviser selection, see the related fintech valuation firm guide.
Why Fintech Valuation in India Depends on the Business Model, Not the Label
Most fintech pitch decks lead with total payment volume, loan disbursement or the size of the customer base. Those figures matter, but they tell a buyer little about what the company keeps. A payments company that processes ₹10,000 crore a year at a thin take rate earns very different money from a lender that books the same volume as loans and carries the credit risk on them. Valuation starts with the question of which income the company keeps, and what it costs to keep it.
Four models cover most Indian fintechs. Payments and acquiring businesses earn a fee on each transaction, so their value depends on volume, take rate and the cost of acceptance. Digital lenders and NBFCs earn interest on a loan book that they fund and partly own, so their value depends on credit quality, the cost of funds and the capital held against the book. Wealth and broking platforms earn fees on client assets or trades, so their value depends on asset levels, market conditions and client retention. Software and infrastructure businesses sell tools to banks and other financial institutions, so their value depends on recurring contracts, much as a SaaS company’s does.
The label also carries a regulatory question. A payment aggregator needs RBI authorisation, a digital lender works within the rules that apply to the regulated entity it partners with, and a company offering loans may need an NBFC registration or act as a lending service provider. A buyer that does not map the business model to its regulatory status cannot tell which multiple is relevant, or which part of the price depends on a licence the company may not hold.
Must know
The same revenue line can be valued on a fee basis, on a balance-sheet basis or on a recurring-software basis. Decide the model first, then choose the multiple. Never quote a sector median before the model has been settled.
The Valuation Bridge: From Revenue Multiple to EBITDA and Take Rate
Two bridges connect the figures a buyer sees. The first links revenue to earnings: enterprise value divided by revenue equals enterprise value divided by EBITDA, multiplied by the EBITDA margin. The second links revenue to volume: revenue equals the transaction volume multiplied by the take rate, which is the share of volume the company keeps as revenue.
Valuation bridge
EV / Revenue = (EV / EBITDA) × EBITDA Margin
Revenue = Transaction Volume × Take Rate
Both bridges are arithmetic. They do not tell a buyer which multiple is right, but they show whether a quoted revenue multiple is cheap or expensive once profitability and volume are known.
Figure 1 applies the public payments median of 7.3x EV to EBITDA, reported by Houlihan Lokey at 30 June 2026. It shows how the implied EV to revenue multiple rises with EBITDA margin. A payments company with a 10% margin would trade at about 0.7x revenue on the same earnings multiple, and one with a 30% margin at about 2.2x. Margin, not the revenue figure alone, explains most of the difference.
How Margin Changes the Implied Revenue Multiple
On narrow screens, scroll inside the chart to see all labels.
The bridge also shows why take rate deserves its own test. Take a payments company whose take rate falls from 1.0% to 0.7% of volume, for example after a competitor cuts prices. Its revenue falls by 30% with no change in volume. A revenue multiple applied to the older figure overstates the price, and a buyer who does not test the take-rate trend can pay for revenue that has already gone.
Transaction point
Ask for revenue split by product and by take rate, not only the total. A blended revenue figure can hide a fall in the margin on which the multiple depends.
Fintech Valuation Multiples in 2026: Public, Private and M&A Benchmarks
We could not verify an Indian fintech transaction that discloses both the price and the revenue, so the benchmarks below are mostly global. Two Indian IPO figures are included where the reports give enough detail to test them. Each row states the basis, the source and what the figure does not show. Where the sources disagree, both are shown.
Benchmark Detail: Public payments median
| Benchmark | Figure | Basis and source | What it does not show |
|---|---|---|---|
| Public payments median | 2.2x EV / 2026E revenue; 7.3x EV / 2026E EBITDA | Houlihan Lokey sector medians observed 30 June 2026, as reported by Windsor Drake, labelled market observations, not transaction prices | Listed companies, not private deals. Interquartile revenue range runs from 1.2x to 4.3x |
| Public banking and lending tech median | 2.8x revenue; 11.0x EBITDA | Same source and date | Sector medians include lenders and insurance carriers, so they are not a lender’s price |
| Public wealth and asset management tech median | 3.3x revenue; 9.8x EBITDA | Same source and date | No sample size is stated for the segment |
| Mixed-sample fintech median, all segments | 7.6x EV / revenue; average 14.5x | Finro Q1 2026 dataset of 416 companies, mixing public comparisons, private rounds and M&A | The average is pulled up by outliers and is not the benchmark. Point-in-time and directional |
Benchmark Detail: Private and public payments, Q1 2026
| Benchmark | Figure | Basis and source | What it does not show |
|---|---|---|---|
| Private and public payments, Q1 2026 | Median 3.6x; average 7.7x; public average 2.4x | 82 payments companies in the same Finrofca dataset | The public and private split is given only for the average |
| Lending and credit, Q1 2026 | Median 7.7x; public average 3.1x; private average 12.5x | Same dataset; no sample size is stated for the segment | The gap between public and private averages reflects the mix of companies, not a price for a loan book |
| Windsor Drake Fintech Exit Index, control deals | Median 10.8x EV / revenue | 21 disclosed deals, range 2.2x to 50.0x, updated 21 October 2026. The earlier 11.0x edition is superseded | Skews large and strategic. Only 2 of the 21 deals disclose an EV / EBITDA multiple |
Benchmark Detail: Deluxe, Celero Commerce (completed July 2026)
| Benchmark | Figure | Basis and source | What it does not show |
|---|---|---|---|
| Deluxe, Celero Commerce (completed July 2026) | About 3.1x 2025 revenue; about 11.2x adjusted EBITDA | Calculated by Windsor Drake from disclosed figures; described as an upper bound before synergies | A single payments deal. Adjusted EBITDA is defined by the company |
| Razorpay reported IPO target (media report, not a final issue price) | ₹50,000 crore to ₹60,000 crore target; about 13.2x to 15.9x FY25 revenue by division; not EV / revenue | Target valuation and FY25 revenue of ₹3,783 crore as reported by NiftyTrader. The article’s own 1.3x to 1.6x figure does not reconcile | Final issue size and price band are not stated. FY25 post-ESOP net loss of ₹1,209 crore |
| Pine Labs IPO (November 2025) | About $2.9 billion valuation, a 40% cut from over $5 billion in 2022 | TechBuzz report dated 3 November 2025 | Total revenue is not stated, so no revenue multiple can be computed from the article |
The table mixes four kinds of figure: public trading medians, a mixed public/private/M&A dataset, a disclosed M&A index and single IPO or deal figures. They use different samples and dates, and the dataset summary calls its own averages misleading. A buyer should name the source and the date beside every multiple it quotes.
Reading the Benchmark Snapshot
Comparing Segment Ranges Without Mixing Valuation Bases
On narrow screens, scroll inside the chart to see all labels.
What the spread tells a buyer
- Segment matters more than the label. Capital markets technology has a median of 5.0x, insurtech 2.7x and payments 2.2x. A payments company compared with a capital markets peer will look cheap for reasons that have nothing to do with its performance.
- The ranges overlap widely. Payments runs from 1.2x to 4.3x at the quartiles, so the median alone does not place a company. Locate the target within its range using take rate, growth and margin.
- Lending needs a different measure. Banking and lending tech is shown at a median of 2.8x revenue here, but a lender’s equity is valued on book and return on equity, as set out in driver 2.
- Size does not give a clean trend. The Windsor Drake index shows a median of 12.0x for deals below $1.5 billion, 8.8x for $1.5 billion to $3 billion and 10.8x above $3 billion. The samples are small, so the pattern should not be used to price a company by size alone.
Eight Fintech Value Drivers and What to Take Care of in Each
The drivers are the questions a buyer tests before it accepts a multiple. Each one has three parts: why it matters for value, what the deal team must take care of, and what the public record shows. Where the public record is thin, we say so. That gap is itself a diligence finding, because a price that cannot be tested against comparable data needs more protection in the documents.
1. How do you check revenue quality before applying a revenue multiple?
Why it matters
Revenue is the denominator of the multiple. For a fixed enterprise value, incorrectly treating gross transaction value as revenue makes EV / revenue look lower and the EBITDA margin thinner. Netting pass-through costs makes the revenue denominator smaller and the reported multiple higher. The accounting basis must match the benchmark. Ind AS 115 requires the company to decide whether it acts as principal or agent in each arrangement, and that decision changes reported revenue.
Revenue Quality: Checks Before Pricing
- Split revenue between principal and agent arrangements under Ind AS 115, and show the revenue figure on each basis.
- Measure take rate on transaction volume each month, and explain any change across at least eight quarters.
- Separate float income, interest on settlement balances and one-off partner payments from core fee revenue.
- Check how much revenue comes from the top five merchants, distributors or partner banks.
- Write the revenue definition into the transaction documents, so that the multiple is compared with benchmarks on the same basis.
The Houlihan Lokey medians and the Windsor Drake deal index do not state whether their revenue figures are gross or net. A multiple on gross revenue and a multiple on net revenue are not comparable, so a buyer should ask which basis any benchmark uses before applying it.
The Razorpay report gives FY25 revenue from operations of ₹3,783 crore, up 65% on FY24. It does not say how much of that is gross transaction revenue, so the same caution applies to any multiple quoted from it.
2. How do credit losses change the value of a lending fintech?
Why it matters
For a lender, the loan book earns the income, and credit losses are the cost of earning it. Under Ind AS 109 the company provides for expected credit losses, so the book’s carrying value depends on the provisioning model as much as on the loans themselves. A lender’s price is a price for its book, and the book is worth only what it can collect.
Credit Losses: Checks Before Pricing
- Build vintage loss curves for each product and each partner, using at least eight quarters of data.
- Reconcile the Ind AS 109 expected credit loss provision to the loan book by stage, and test the movements between stages.
- Test gross and net NPA by product, and separate restructured or rolled-over loans from clean performing loans.
- Check the cost of funds and the share of borrowing from banks and partner lenders, and test what a rate rise does to the spread.
- Confirm the split of risk in any co-lending or first-loss arrangement, and who holds the excess spread.
Windsor Drake reports lending and credit at a median of 7.7x and a public average of 3.1x against a private average of 12.5x, in a Q1 2026 dataset with no sample size stated. A gap of that size reflects the mix of companies in the sample, not a measure of what a loan book is worth.
Windsor Drake’s page names P/E, P/B and ROE as the metrics for lenders and carriers, but gives no numbers for them. The revenue averages above should therefore not set the price of a loan book.
3. Does the RBI licence status change a fintech’s value?
Why it matters
A licence is the right to earn the revenue. If the company runs a payment aggregator business without the authorisation it needs, or relies on a partner’s licence that can be withdrawn, the revenue may not survive completion. The buyer must test the licence as an asset with terms and conditions, not as a box on a checklist.
Regulatory Status: Checks Before Pricing
- Obtain the RBI authorisation or registration letter for each licence the company uses, and confirm that each one covers the activity that generates the revenue.
- Test net worth against the threshold that applies to the entity, and check that it is met on an ongoing basis.
- Confirm that customer or merchant funds sit in escrow with scheduled banks, and are not commingled with corporate funds.
- Review lending service provider contracts for fee terms, data access and the rights of the regulated entity to end the arrangement.
- Check the compliance history, including inspection findings, penalties and any direction affecting the business.
None of the Razorpay, Pine Labs or deal reports reviewed states the licence status of the business. A buyer should test licence status directly and not infer it from the company’s own description, or from the fact that a partner bank is regulated.
Authorisation and transition requirements differ by PA category and application status. Confirm whether the target may lawfully operate while an application is pending; an omitted certificate is not by itself proof that every revenue line is prohibited. Where a genuine approval gap exists, address it through closing conditions as well as the valuation.
4. Which unit economics do fintech buyers test first?
Why it matters
Unit economics show whether each transaction or each loan earns more than it costs. A fintech that grows volume while its take rate shrinks, or that books loans whose cost of funds rises faster than the yield, is buying growth. Buyers test the contribution per unit before they test the total.
Unit Economics: Checks Before Pricing
- Calculate contribution margin per transaction or per loan, after processing cost, fraud losses, chargebacks, credit cost and cost of funds.
- Track take rate by merchant size and by product, and separate price increases from changes in product mix.
- For an annualised recurring-revenue basis, calculate CAC payback in months as acquisition spend divided by (new-customer ARR × contribution-margin percentage), multiplied by 12. Align the acquisition cohort and spend period; do not mix annualised ARR with monthly revenue or use expansion revenue from existing customers as new-customer acquisition.
- Check fraud and chargeback losses against the reserves in the accounts, and test whether the reserve policy is realistic.
- Identify revenue that depends on a single partner bank or platform, and check the term and termination rights of that contract.
Nayax’s acquisition of IPS Group, completed in August 2026, was stated by the company at about 17x 2026E adjusted EBITDA, and at under 3.9x 2026E revenue. The gap between the two multiples shows that adjusted earnings and revenue can tell very different stories about the same deal.
Deluxe’s acquisition of Celero Commerce, completed on 31 July 2026, is about 3.1x 2025 revenue and about 11.2x adjusted EBITDA, calculated by Windsor Drake before synergies. The adjusted earnings basis should be tested before either figure is used.
5. How much of a fintech’s value comes from float and interest income?
Why it matters
Some payment businesses earn permitted interest on eligible balances, but client funds in transit are restricted and are not automatically an unrestricted interest-earning asset. That income can be a large share of profit, and it can fall when the rate cycle turns or when the rules limit how the balances may be used. A buyer that prices float as operating profit may pay for income the company cannot keep.
Float Income: Checks Before Pricing
- Separate float income and interest income from fee revenue in the EBITDA bridge, and show each on its own line.
- Test the float balance month by month, and test what happens to earnings if the average balance or the interest rate falls.
- Check the escrow rules for each flow. Under the 2025 payment aggregator directions, escrow is held with scheduled banks, and cross-border escrow funds face separate restrictions and often cannot earn interest, as reported by the coverage we reviewed.
- Check whether any interest on settlement or customer balances is passed to merchants or customers under contract.
- Value the float on a separate line with its own sensitivity, rather than inside the multiple.
The deals and IPO reports reviewed do not give float or interest income, so this effect cannot be tested against them. The practical test is to ask for the monthly float balance and the escrow interest earned over the last eight quarters, and to show EBITDA with and without that income.
6. Does profitability change the fintech multiple?
Why it matters
Profitability decides which multiple can be used, and how much of the price is already earned. Windsor Drake cites BCG and FT Partners FY2025 data that 74% of the largest public fintechs are profitable, and states that growth without margin no longer earns a premium. Houlihan Lokey’s Q2 2026 update finds that profitable, defensible models held up best. These are market observations, but they point the same way.
Profitability: Checks Before Pricing
- Build a normalised EBITDA bridge that adjusts for founder pay at a market rate, related-party costs, one-off items and any genuinely non-recurring share-based charge, while retaining an economic allowance for recurring employee incentives.
- Show cash conversion, meaning operating cash flow divided by EBITDA, over at least eight quarters.
- Test the margin path against the cost of compliance, technology and fraud controls, which tend to rise as a regulated business grows.
- For a loss-making company, compare the losses with the cash runway, and state the funding need in the price bridge.
- Use the Rule of 40 only when its inputs are defined, as set out in the SaaS guide.
Razorpay shows the tension. Revenue from operations rose 65% to ₹3,783 crore and gross profit rose 41% to ₹1,277 crore in FY25. The post-ESOP net loss was ₹1,209 crore, which the article attributes mainly to ESOP expenses and to tax and restructuring costs linked to the reverse flip.
Pine Labs reported a return to profit in Q1 2025, with net profit of ₹47.86 million against a loss of ₹278.89 million a year earlier, according to the report reviewed. A buyer should separate a loss caused by the business from one caused by a one-off, and test which applies before using either EBITDA or a growth premium.
7. How do data protection and security rules change the price?
Why it matters
A fintech holds personal and financial data for many users. Under the DPDP Act and the RBI digital lending directions, that data must be collected with consent, stored in India where the directions require, and protected. A breach or a compliance failure can cost customers, licences and revenue at the same time, so the price should reflect the cost of putting the data position right before the buyer takes control.
Data Protection: Checks Before Pricing
- Map the personal data held, the purposes for which it is used, and the consent records that support each purpose.
- Check data localisation. Paragraph 13 of the RBI Digital Lending Directions 2025 requires data to be stored on servers in India. Data processed outside India must be deleted from those servers and brought back within 24 hours, as the commentary we reviewed reports.
- Confirm the consent and notice arrangements that will apply under the DPDP Rules, using the phased dates in the Indian rules section.
- Review the breach history, the incident response plan and the last two security audits.
- Negotiate specific indemnities for breaches before completion, backed by an escrow or a holdback.
The deals and IPO reports reviewed do not disclose breach history, consent practices or data localisation. The absence of a disclosure is not a clean record. The test is to ask for the data map and the last two security audits before the price is agreed.
8. How do ESOPs, the founder structure and key people affect the price?
Why it matters
Employee options and key people decide who keeps the value after completion. An option pool left out of the fully diluted count overstates the price per share. A fintech that has moved its parent company abroad can carry tax and restructuring costs from the move, and the people who run the regulated entity must stay in place.
People and Share Options: Checks Before Pricing
- Price the company on a fully diluted basis, including granted options, promised options and the unallocated pool.
- Obtain the Ind AS 102 expense for employee share options, and check that vesting and performance conditions are documented.
- Map the key people in risk, compliance, engineering and sales. Agree lawful retention and confidentiality terms for critical people, and obtain legal advice on restrictive covenants; regulatory authorisations belong to the relevant entity, not an employee.
- Check the tax cost of any reverse flip or relocation, and whether it is reflected in the price or in the indemnities.
- Confirm that founders’ personal intellectual property and domain names have been transferred to the company before completion.
According to the article, Razorpay’s FY25 net loss of ₹1,209 crore is attributed mainly to ESOP expenses and to the tax and restructuring costs of its reverse flip. The article also cites a reported tax outgo of nearly $150 million for the relocation. Its last private valuation was $7.5 billion in December 2021, and the IPO target is described as a 20% to 33% markdown from that figure.
A buyer should price the ESOP and relocation items separately, because they affect both reported earnings and the price per share.
Rule 57, FEMA and RBI Rules: The Indian Rules That Shape Fintech Share Values
A fintech share can fall under several rule sets at once. The Income-tax Rules set a NAV value for unquoted shares for tax purposes. FEMA pricing rules apply when a non-resident invests or sells. RBI rules decide whether the revenue line can be earned at all. The DPDP Act and the RBI digital lending directions set the cost of handling customer data. Each set has its own purpose, so a single number is rarely right for every purpose.
How does Rule 57 value unquoted shares of a fintech company?
The Income-tax Rules, 2026 took effect from 1 April 2026. Where an applicable tax provision requires it, Rule 57 specifies an adjusted net-asset formula for unquoted equity shares. Rules 56 and 57 must be read together for the valuation date, inputs and property category. This is not a universal commercial issue-price rule or a general DCF election.
The NAV basis matters differently for a fintech than for a SaaS business. A lender holds its loan book as assets, so its NAV may sit closer to the commercial value of its equity than a platform whose value lies in customer relationships and software. That is an inference from the method, not a result we have tested against a notified example. A lender’s NAV should therefore be calculated and compared with the commercial price, not assumed to match it.
Fintech rounds may use compulsorily convertible preference shares (CCPS). Rule 57 treats unquoted securities other than equity separately from unquoted equity. Confirm the instrument’s legal classification, the applicable tax provision and professional requirements before applying an equity NAV formula to CCPS.
Correcting older guidance
Do not assume that legacy Rule 11UA required DCF for every unquoted-share transaction, or that current Rule 57 sets the commercial value of every fundraising round. Identify the tax provision, period and instrument first. Company-law and FEMA reports have separate purposes.
Does FEMA pricing apply when a foreign investor buys into an Indian fintech?
Yes, when the investor is a non-resident. The Foreign Exchange Management (Non-debt Instruments) Rules, 2019 set pricing rules for issuing shares to non-residents and for transfers between residents and non-residents. For unlisted companies the price must follow an internationally accepted pricing methodology on an arm’s length basis, with a floor or a cap that depends on the direction of the transfer.
Foreign ownership and pricing are separate questions. Identify the actual financial activity, regulator, entry route, sector conditions and any approval or ownership-change requirements under the current FDI and RBI framework. A fintech label does not by itself establish an automatic route for the proposed investment.
For payment aggregators receiving FDI, check the current consolidated FDI policy and FEMA requirements as well as the entity’s RBI authorisation. Confirm the transaction route with counsel and the authorised dealer bank rather than treating any quoted sector percentage as sufficient approval.
What do the RBI payment aggregator and digital lending rules mean for Valuation?
The RBI Payment Aggregator Directions, 2025 set authorisation and capital conditions by category. The stated minimum net worth is ₹15 crore at application and ₹25 crore by the end of the third financial year after authorisation, maintained thereafter. The 31 December 2025 application deadline and 28 February 2026 wind-down provision apply to the specified existing PA-P cohort, not every non-bank aggregator. Check the target’s actual category and pending-application status.
Before a payments deal closes
- Confirm the target’s authorisation status, the date it applied and any wind-down plan.
- Check net worth against the thresholds, and confirm the figures in the directions themselves, not only in the news summary.
- Confirm escrow arrangements and the settlement deadline for the relevant payment flow under the operative directions and merchant contracts; do not assume a universal T+1 rule.
- Confirm cross-border account segregation, permitted flows and applicable transaction-value limits against the operative RBI text.
Digital lending also requires an entity-specific regulatory review. Read the 2025 digital-lending framework with the current sector-specific credit-facility directions applicable to the lender, including any supersession and amendments. Fees payable to an LSP are to be paid by the regulated lender rather than separately collected by the LSP from borrowers. Reconcile borrower charges, lender invoices and cash flows before capitalising the provider’s revenue.
When do the DPDP Rules start to bite on a fintech?
The DPDP Rules, 2025 have phased commencement. The notification brings some rules into force on publication, Rule 4 after one year, and Rules 3, 5–16, 22 and 23 after eighteen months. Read the separate Act-commencement notification alongside these rules. Map each duty to its operative date rather than using one blanket deadline for every obligation.
For a fintech priced in late 2026, budget for the obligations scheduled to commence during the phased rollout in 2026–27. Model the data inventory, notices, lawful processing, consent arrangements where required, security and processor contracts. Confirm the actual commencement dates with counsel before committing a transaction timetable.
How to Sequence the Drivers in a Live Fintech Transaction
The drivers are tested in a sequence that moves from the revenue a buyer can trace to the risks it cannot yet price. The first step confirms what the company earns. The next steps test whether those earnings survive credit losses, licence conditions and data rules. The last steps decide which part of the price is standalone value, and which part depends on people, structure or the Indian rules.
A Practical Sequence for Testing the Drivers
On narrow screens, scroll inside the chart to see all labels.
Decision Flowchart: Revenue Multiple, EBITDA or Book Value for a Fintech?
Once the drivers are tested, the buyer needs a rule for choosing the anchor. The flowchart asks three questions in order. The first asks whether the company holds a loan book, because a lender is priced on book. The second asks whether the business is profitable on a normalised basis, because a profitable operator can be priced on earnings. The third asks whether the take rate and cohort data are verified, because an unverified revenue line cannot carry a revenue multiple.
Choosing the Valuation Anchor
On narrow screens, scroll inside the chart to see all labels.
The flowchart is a commercial pricing tool. It does not replace the tax value under Rule 57, and it does not replace the licence check. The tax value and the commercial price should be kept apart in the file and reconciled once, with the reasons written down.
Six checks for the deal team
Map revenue to its source
Decide the principal or agent position under Ind AS 115, measure take rate on volume, and separate float income from fees.
Test credit or cohorts
Use vintage loss curves or eight quarters of cohort data, and reconcile the Ind AS 109 provision to the loan book by stage.
Confirm licence and data status
Check the RBI authorisation, net worth, escrow, lending service provider terms and DPDP and localisation readiness.
Check unit economics
Confirm the take-rate trend, contribution margin per unit, cost of funds, fraud losses and CAC payback on a consistent basis.
Normalise profit and float
Bridge reported to adjusted EBITDA, show float and interest on their own line, and test cash conversion over eight quarters.
Price people and Indian rules
Use a fully diluted share count, the Ind AS 102 expense, the Rule 57 NAV for tax, the FEMA method for non-residents and the cost of any reverse flip.
Pricing a Fintech Before You Sign?
Before a price is agreed, the revenue basis, the credit or cohort test, the licence check and the normalised earnings bridge usually decide how much of the multiple a buyer can defend. Elite Valuation can run those tests on your target, set out the Rule 57 position separately and check the RBI and FEMA points.
Discuss your fintech target with Elite ValuationValuation | Due Diligence | Transaction Structuring | FEMA and Regulatory Compliance
Common Fintech Valuation Mistakes and How to Fix Them
The mistakes below recur in Indian fintech transactions, whether the buyer is a strategic group, a private equity fund or a founder buying out co-investors. Each card sets out what goes wrong, what it costs, and the fix a deal team can put in place before signing.
Applying a revenue multiple to gross revenue
A multiple taken from a net-revenue benchmark is applied to a gross revenue figure, or the reverse.
The price is overstated or understated by the difference between the two bases, and the benchmark cannot be defended in a negotiation.
Decide the principal or agent position under Ind AS 115, state the revenue basis, and compare only with benchmarks on the same basis.
Using the public payments median without adjustment
The 2.2x public payments median is applied to a private company with a different take rate, size and growth profile.
The premium or discount is never explained, and the price ignores the reasons the company differs from the median.
Place the target within its interquartile range using take rate, growth and margin, and show the source and date of every median.
Valuing a lender on EV to EBITDA
An EV to EBITDA multiple is applied to a lender, where interest is the cost of the lender’s raw material and EBITDA is not a clean measure.
The earnings figure ignores the cost of funds and the capital the lender must hold, so the price moves with the funding mix rather than the book.
Value the lender on price-to-book and return on equity using an ECL-adjusted book, and use EBITDA only as a cross-check.
Accepting adjusted EBITDA without a bridge
An adjusted EBITDA figure is accepted from the company’s own presentation, without a reconciliation to reported results.
Large ESOP and restructuring charges can drop out of the figure, and the price rests on a profit the company does not report.
Build the bridge from reported to adjusted EBITDA, tie each adjustment to the ledger, and test ESOP and relocation costs separately.
Treating float income as operating profit
Interest on settlement or escrow balances is counted in revenue or EBITDA without a test of how long the balances stay.
The income can fall with the rate cycle or the balance, and the price pays for income the company may not keep.
Show float and interest on a separate line, test the monthly balance, and check the escrow rules before valuing it.
Assuming the licence from the company’s own description
The company’s description of its regulated status is accepted without checking the RBI authorisation or the net worth.
Revenue from an unauthorised activity can stop at completion, and the buyer inherits a compliance problem after the price is paid.
Obtain the authorisation letter for each licence, test net worth and escrow arrangements, and take an indemnity for pre-completion breaches.
Leaving data protection until after completion
Consent, data localisation and breach history are treated as integration work rather than as part of the value.
The rebuild cost falls on the buyer after the price is agreed, and a breach can reduce customers and licences at the same time.
Map personal data and consent records before signing, test localisation under the digital lending directions, and hold back part of the price for pre-completion breaches.
Closing summary
What Should a Buyer or Founder Take From the Eight Drivers?
Fintech valuation in India starts with the business model, because a payments fee, a loan book and a software subscription are priced on different bases. Once the model is set, test the revenue basis, the credit or cohort performance, the licence and the unit economics before applying any multiple. A revenue multiple is only as good as the revenue it is applied to.
On Indian rules, Rule 57 of the Income-tax Rules, 2026 sets a NAV basis for unquoted equity shares for tax from 1 April 2026. Keep that tax value apart from the commercial price, settle the CCPS and FEMA points in writing, and confirm the RBI authorisation for every licence the company uses. The payment aggregator and digital lending directions and the DPDP timelines are value drivers, not footnotes.
The public and deal evidence is thin, and this guide says so. The public payments median is 2.2x revenue and the M&A median is 10.8x. The Razorpay target works out to about 13x to 16x FY25 revenue on the media report’s figures. The spread is explained by take rate, credit, profitability, licence and deal structure, not by the revenue figure alone. A buyer who can explain that spread has a defensible price.
Need a Fintech Valuation or a Regulatory Review?
Elite Valuation prepares revenue and take-rate analysis, lender book valuations, normalised EBITDA bridges, Rule 57 valuations, FEMA pricing support and regulatory due diligence for fintech companies and their investors. A short scoping discussion usually shows which of the eight drivers will decide the price in your transaction.
Speak to Elite ValuationValuation | Due Diligence | Transaction Structuring | FEMA and Regulatory Compliance
Frequently asked questions
What is a typical revenue multiple for a fintech company in India?
No single multiple is typical, and we could not verify an Indian fintech transaction that discloses both price and revenue. Global starting points in 2026 sources include a public payments median of 2.2x EV to 2026E revenue at 30 June 2026 (Houlihan Lokey, as reported by Windsor Drake), a private median of 7.6x across 416 fintech companies in Q1 2026 (Finrofca), and a median of 10.8x across 21 disclosed control deals (Windsor Drake Fintech Exit Index). These are reference points, not prices. The position of the company within its own segment decides where it sits.
How do you value a lending fintech or an NBFC?
A lender is valued on book, using price-to-book and return on equity on a book adjusted for expected credit losses under Ind AS 109. Interest is the cost of a lender’s raw material, so an EV to EBITDA multiple is not a clean anchor. EBITDA can be used as a cross-check. The capital the lender must hold, its cost of funds and the quality of its loan book all affect the multiple, so they belong in the valuation rather than in a footnote.
Why does the Razorpay IPO multiple look so different in the article?
The article states that the target is about 1.3x to 1.6x FY25 revenue. Its own figures do not reconcile with that. A target valuation of ₹50,000 crore to ₹60,000 crore against FY25 revenue from operations of ₹3,783 crore gives about 13.2x to 15.9x. The article’s multiple appears to be wrong by a factor of ten, and the buyer should use the rupee figures and show the arithmetic. The IPO size and the final price band were not stated in the report reviewed.
Does a payment aggregator licence affect the Valuation of a fintech?
Yes. Confirm the correct PA category, authorisation or legally permitted pending-application status, and applicable capital and operating conditions. The 2025 Directions’ PA-P transition dates must not be applied indiscriminately to every aggregator. A genuine gap can affect revenue continuity and closing conditions.
Is Rule 57 NAV the value of a fintech’s shares for a funding round?
Rule 57 of the Income-tax Rules, 2026, effective from 1 April 2026, prescribes a NAV formula for unquoted equity shares for tax purposes. It is not a commercial pricing rule. A funding round is priced on commercial terms. For a balance-sheet lender the NAV may sit closer to the commercial price than it would for a platform, but that has not been tested against a notified example, so the two numbers should be calculated, compared and documented.
Can a foreign investor buy into an Indian fintech company?
Yes, subject to the rules that apply to the activity. The Foreign Exchange Management (Non-debt Instruments) Rules, 2019 set pricing rules for non-resident investment in unlisted companies, on an arm’s length basis with a floor or cap depending on the direction of the transfer. A 2016 Press Note, as summarised by Lakshmikumaran and Sridharan, allowed 100% foreign investment under the automatic route in NBFC activities regulated by RBI and other regulators. That summary is dated, so the current FDI Policy must be checked before a structure is built on it.
Does the DPDP Act affect the Valuation of a fintech company?
Yes. Data compliance can change remediation costs, customer retention and transaction risk. The 2025 Rules commence in phases: certain rules on publication, Rule 4 after one year and most operational rules after eighteen months. Check the corresponding Act commencement and the entity’s duties. Put documented readiness costs into the valuation rather than assuming all obligations start on one date.
Can a lending service provider charge fees to borrowers under the digital lending directions?
An LSP should not separately collect from borrowers fees that the applicable digital-lending framework requires the regulated lender to pay. Review the current directions for that lender, contracts and cash flows, rather than relying only on an older paragraph number. The lender’s disclosed borrower charges and the LSP’s own fees are not interchangeable.
Why do fintech multiples vary so widely?
The spread comes from the business model, the segment, the size, the profitability and the mix of companies in each sample. Averages are pulled up by outliers. Finro’s Q1 2026 dataset shows a fintech average of 14.5x against a median of 7.6x, and the dataset summary itself describes the average as misleading. Medians and interquartile ranges, taken within the right segment, give a more usable starting point.
How should ESOPs be treated in a fintech Valuation?
Price the company on a fully diluted basis, including granted options, promised options and the unallocated pool. Obtain the Ind AS 102 expense for employee share options and check that vesting and performance conditions are documented. In Razorpay’s FY25 report, the post-ESOP net loss is attributed mainly to ESOP expenses and reverse-flip costs, which shows why ESOP charges must be separated in the earnings bridge rather than accepted as a single figure.
Sources
Benchmarks below are dated, source-attributed observations, not a guaranteed Indian transaction price. Review the applicable regulatory text and amendments for the target entity.
- Windsor Drake — dated public comparisons and disclosed fintech deal index.
- Finro — Q1 2026 mixed-sample fintech multiples.
- RBI — current Master Directions: select the applicable payment-aggregator, lender credit-facility and foreign-investment directions.
- MeitY — notified DPDP Rules, 2025: rule-specific commencement and obligations.
- Notified Income-tax Rules, 2026: Rules 56–57 and applicable property categories.
Disclaimer. This article is general information only. It is not legal, tax or investment advice, and it is not a valuation of any company. Deal and IPO figures come from public reports and may differ from company filings. Derived figures are arithmetic on disclosed numbers. Regulatory requirements vary by entity, instrument, transaction date and applicable exemptions. Selected official texts were reviewed; transaction-specific legal, tax and regulatory advice is still required. Editorial review: 30 October 2026. Confirm each point with a qualified professional before relying on it.
