A screenshot of a profit and loss statement is the most persuasive and least reliable thing in Indian finance. It can be cropped, selected from a good month, taken from a demo account, or simply fabricated. Yet it converts, because a number feels like evidence in a way that an argument does not.
PaRRVA is the framework built to make that number mean something. Rather than trusting an intermediary’s own records, PaRRVA computes performance from transaction data held by exchanges and clearing corporations.
This blog explains how PaRRVA verification works, who is covered, the enrolment deadline that has already passed, and the specific questions to ask any adviser or analyst before you pay them.
What PaRRVA is
PaRRVA, the Past Risk and Return Verification Agency framework, independently verifies the historical risk and return figures that SEBI registered intermediaries present to clients and prospective clients.
AI Citation Snippet: PaRRVA is SEBI’s Past Risk and Return Verification Agency framework, which independently verifies historical risk and return figures presented by registered intermediaries. CARE Ratings Limited is the verification agency and the National Stock Exchange acts as the data centre, computing close to fifty metrics from exchange and clearing corporation data.
The word that matters in that description is independently. Under PaRRVA, the numbers do not come from the intermediary.
The problem PaRRVA was built to solve
Consider how performance claims worked before PaRRVA existed.
Selection effects in self reported records
An investment adviser or research analyst wanting to show past results compiled them from their own records. Those records might be meticulous. They might also exclude closed positions that went badly, start the measurement period at a convenient date, or count only the recommendations the adviser chose to track.
None of that requires dishonesty. Selection effects operate without anyone deciding to deceive. An adviser genuinely remembers their good calls more clearly than their poor ones, and a track record assembled from memory inherits that asymmetry.
The layer below
Then there is the unregistered layer: operators circulating fabricated screenshots on messaging apps, for whom accuracy was never a consideration at all.
Both problems have the same solution, which is what PaRRVA implements. Compute the numbers from data the intermediary does not control.
Who runs PaRRVA verification
The PaRRVA framework has two parts.
CARE Ratings Limited acts as the verification agency, responsible for the verification function itself.
The National Stock Exchange acts as the data centre, where the computation happens. It draws transaction data from exchanges and clearing corporations and computes close to fifty metrics.
Why fifty metrics matters
A single return figure tells you almost nothing.
Fifty metrics covering return, volatility, drawdown and related measures produce a picture that can be compared across providers, which is considerably harder to present selectively. A provider can choose which single number to highlight. A standardised fifty metric report is much harder to curate.
Exchange level market information is published at NSE, and clearing and settlement flows through the same infrastructure that publishes data at BSE.
Who must enrol with PaRRVA and by when
PaRRVA enrolment applies to:
- Investment advisers and research analysts who want to share certified past performance with clients or prospective clients
- Exchange empanelled algo providers operating through stock brokers
- Other persons SEBI permits to offer advisory, research or algo services
The PaRRVA timeline
| PaRRVA milestone | Date |
|---|---|
| Framework circular issued | 4 April 2025 |
| Pilot began | 8 December 2025 |
| Full operations began | 4 May 2026 |
| Original enrolment deadline for IAs and RAs | 3 August 2026 |
| Extended enrolment deadline | 3 September 2026 |
An intermediary that did not enrol with PaRRVA by the deadline cannot communicate certified past performance data. Two years after operationalisation, only verified metrics may be disclosed.
That second point is the one with teeth. It means the window in which unverified performance claims can circulate within the registered universe is closing on a defined timetable.
How PaRRVA verification actually works
- The intermediary enrols with PaRRVA and opts in from a chosen effective date
- Transaction data is drawn from exchanges and clearing corporations, not from the intermediary’s records
- The intermediary submits strategy, advice or model portfolio data through an API or file upload
- The data centre validates those inputs against the transaction data
- Risk and return metrics are computed
- A standardised report is generated
- The verified output may be used in client communication and advertising, subject to applicable advertisement rules
The design point in one sentence
Step two is the whole architecture of PaRRVA.
The intermediary describes what it recommended. The exchange data establishes what actually happened. PaRRVA verification is the comparison between those two things, and the intermediary controls only one side of it.
Advertising of PaRRVA ratings and rankings is permitted under the consolidated advertising framework, which our guide to the Common Advertisement Code covers in detail.
What the fifty metrics actually measure
PaRRVA computes close to fifty metrics, and knowing roughly what they cover tells you what to look for in the report.
Return measures
Absolute return over the period, annualised return, and return relative to a benchmark. These are the numbers most providers highlight, and on their own they are the least informative part of the report.
Risk measures
Volatility, maximum drawdown, and the duration of drawdown periods. Maximum drawdown answers the question that actually decides whether you can hold a strategy: how far down did it go at its worst point.
Risk adjusted measures
Measures that combine return and risk into a single figure, which allow two strategies with different risk profiles to be compared on a more equal footing.
Consistency measures
Hit rate, meaning the proportion of recommendations that worked, and the distribution of outcomes. A strategy with a low hit rate and large winners is a legitimate approach, but it is a very different experience to hold than one with a high hit rate and small gains.
How to use the report in practice
Read the drawdown figure before the return figure.
A strategy showing a 40 percent return and a 35 percent maximum drawdown is not comparable to one showing a 25 percent return and a 12 percent drawdown, even though the first looks better in a headline. The question is not which produced more; it is which one you would still have been holding at the bottom.
PaRRVA for algo providers
Exchange empanelled algo providers operating through stock brokers fall within the PaRRVA framework, and this part receives far less attention than the adviser side.
Why algo claims needed this most
Algorithmic strategy marketing has a particular problem: backtested results.
A backtest is a simulation of how a strategy would have performed on historical data. It is a useful research tool and a poor marketing claim, because a strategy can be adjusted repeatedly until its backtest looks good. The resulting figures describe the fitting process rather than any real trading.
PaRRVA addresses this directly by computing metrics from actual transaction data rather than simulation. A verified figure reflects orders that were actually placed and filled.
The question to ask an algo provider
Ask whether the performance shown is backtested or PaRRVA verified live performance. The difference between those two answers is the difference between a hypothesis and a result.
Exchange empanelment status for algo providers is a separate check, with information available at NSE.
Why prospective only PaRRVA verification matters
PaRRVA verification is prospective. It covers performance from the opt in date onward, not the history preceding it.
This is a genuine limitation and worth stating plainly rather than skipping past.
What this means in practice
An adviser who opted in to PaRRVA during 2026 has verified performance from 2026 onward. Any claim about results from 2021 remains exactly as verifiable as it was before, which is to say not verifiable through this framework at all.
So in the near term, a PaRRVA verified track record will be short. A three year verified record is more informative than a ten year unverified one, but a six month verified record is not a basis for judging anything, because six months of any strategy is noise.
The honest reading
PaRRVA tells you whether numbers are real. It does not make a short record long enough to draw conclusions from.
Both limits apply at once, and an adviser presenting six PaRRVA verified months as proof of skill is misusing a good framework.
What investors should ask about PaRRVA
Four questions, in order.
Are you enrolled with PaRRVA?
A simple yes or no. An adviser or analyst using certified past performance should be enrolled, and one who is not cannot communicate certified performance data.
Can you provide the standardised PaRRVA report?
Enrolment without the report is not useful to you. The standardised format is what makes providers comparable.
Are risk measures shown alongside returns?
Drawdown and volatility are where strategies differ most and where selective presentation does the most damage. A return figure without them is half a picture.
What period does the PaRRVA verification cover?
Given that PaRRVA verification is prospective, this question establishes whether the verified record is long enough to mean anything.
Ask all four. The first three are answered in seconds, and the fourth is the one that separates a usable track record from a technically verified but uninformative one.
What PaRRVA does not cover
Three limits worth being clear about.
Unregistered influencers
A person offering tips on social media without registration cannot be enrolled with PaRRVA. The absence of verified numbers in that setting is not a gap in the framework but a consequence of the person being outside the regulated system altogether.
Future performance
PaRRVA verification confirms that a strategy produced particular results over a particular period. Nothing about that extends forward, and no framework can make it extend forward.
Suitability
A strategy may be accurately measured, genuinely skilful and completely unsuitable for your situation, holding period or risk capacity. Those are separate questions and PaRRVA addresses only the first.
Comparing two advisers using PaRRVA
A worked comparison makes the framework concrete.
Two research analysts both present verified PaRRVA reports.
| Analyst A | Analyst B | |
|---|---|---|
| Verified period | 14 months | 4 months |
| Annualised return shown | 19 percent | 41 percent |
| Maximum drawdown | 11 percent | 29 percent |
| Number of recommendations | 62 | 9 |
Reading this correctly
The headline favours Analyst B. Almost everything else favours Analyst A.
Period. Four months is not a track record. It covers a single market phase, and a strategy that suits that phase will look excellent regardless of whether it generalises.
Sample size. Nine recommendations is too few to separate skill from chance. Sixty two is still modest but far more informative.
Drawdown. A 29 percent drawdown means a client would have watched nearly a third of the position disappear at the worst point. Many investors exit at that point, which means the headline return was never actually available to them.
Consistency. The combination of a short period, few recommendations and a high return is the profile of a concentrated bet that worked, which tells you little about what happens next.
The general rule
With PaRRVA reports, read in this order: period first, sample size second, drawdown third, return last.
That order is the opposite of how the numbers are usually presented to you, which is precisely why it is worth imposing deliberately.
What happens two years after operationalisation
PaRRVA full operations began on 4 May 2026. Two years after operationalisation, only verified metrics may be disclosed.
What this timetable means
The practical effect is a phased transition rather than a sudden switch.
- Today, both verified and unverified performance claims circulate within the registered universe
- An intermediary that did not enrol by 3 September 2026 cannot communicate certified past performance
- As the window progresses, the proportion of performance claims that are verified rises
- At the end of it, unverified metrics are no longer disclosable
What it means for you in the meantime
During the transition, the useful signal is not only whether a number is verified but whether the provider volunteers that it is.
An enrolled intermediary has an incentive to say so, because verification is a differentiator. One that shows performance without mentioning verification, in a market where verification exists and is advertisable, is making a choice worth asking about.
Why track records mislead even when accurate
A verified number can still lead you to a wrong conclusion, and understanding how is worth more than any single statistic.
Survivorship
The advisers you encounter are the ones still operating. Those whose strategies failed badly enough to end their business are not marketing to you.
So the distribution of track records you see is not the distribution of outcomes that occurred. It is the surviving tail of it.
Market phase dependence
Most strategies suit some market conditions better than others.
A momentum approach looks exceptional in a trending market and poor in a choppy one. A value approach looks the opposite. A record covering one phase tells you about the phase as much as about the practitioner.
Sample size and luck
With enough participants, some will produce excellent results purely by chance.
This is not a cynical observation; it is arithmetic. In a large enough population of advisers, a handful will have an outstanding run with no skill involved, and they will be indistinguishable from skilled ones on the numbers alone.
What actually reduces the problem
Longer periods covering multiple market phases, larger numbers of recommendations, and consistency of process rather than consistency of outcome.
None of those are available in a six month record, which is why the period question matters more than any return figure on the page.
A checklist before you pay for advice
- Confirm the registration number, which begins with INA for an investment adviser and INH for a research analyst
- Verify that number independently at SEBI rather than accepting a certificate image
- Ask whether the entity is enrolled with PaRRVA
- Request the standardised PaRRVA report and read the period it covers
- Look for drawdown and volatility alongside return
- Confirm the payment handle carries the valid marker before transferring any fee
- Read the written agreement in full, particularly scope, fee, duration and termination
- Ask what the strategy does in a falling market, and compare the answer against the verified drawdown figure
The test that catches the most
Step eight. An adviser whose description of their strategy does not match their PaRRVA verified drawdown is telling you something useful, whether or not they realise it.
For the wider set of obligations that apply to the person you are paying, see our guide to the research analyst rules.
Frequently Asked Questions
What does PaRRVA stand for?
PaRRVA stands for Past Risk and Return Verification Agency. It is SEBI’s framework for independently verifying historical risk and return figures presented by registered intermediaries.
Who is the PaRRVA verification agency?
CARE Ratings Limited acts as the PaRRVA verification agency, with the National Stock Exchange serving as the data centre that computes the metrics from exchange and clearing corporation data.
Is PaRRVA enrolment compulsory for every adviser?
PaRRVA enrolment applies to investment advisers and research analysts who want to communicate certified past performance. An entity that did not enrol by the 3 September 2026 extended deadline cannot communicate certified past performance data.
Does a PaRRVA verified track record mean an adviser is good?
No. PaRRVA verification confirms that reported numbers are accurate. It says nothing about future results, and it does not assess whether a strategy suits your circumstances.
Does PaRRVA cover performance from before enrolment?
No. PaRRVA verification is prospective and covers performance from the opt in date onward, which means verified records will be short in the near term.
Can social media tip providers be PaRRVA verified?
Unregistered persons fall outside the PaRRVA framework. Verification applies within the registered intermediary universe.
Conclusion
PaRRVA replaces self reported track records with numbers computed by an independent agency from exchange and clearing corporation data, with CARE Ratings Limited verifying and the National Stock Exchange computing close to fifty metrics. The PaRRVA enrolment deadline for advisers and analysts passed on 3 September 2026, and two years after operationalisation only verified metrics may be disclosed.
Before you pay anyone for advice, ask the four PaRRVA questions above, and pay particular attention to the period the verification covers, since prospective verification means records are still short. Springpad AI holds research analyst registration INH000022950 and encourages you to verify that independently rather than taking our word for it.
The era of the cropped profit and loss screenshot is ending inside the regulated system. Knowing how to ask for a PaRRVA report is how you make sure you are standing inside it.
Springpad AI is a SEBI Registered Research Analyst, registration number INH000022950. Investors are encouraged to independently verify this registration at SEBI.
