Guide · Mutual funds · Updated 2026-09-18

Which NAV you get is decided by when your money reaches the scheme, not by when you pressed submit

Most investors carry a mental model built before 2021. It says there is a deadline, the deadline is three o'clock, and an instruction stamped before it buys that day. That model was correct for small purchases for about fifteen years, it is comfortable, and it is no longer how anything works.

The operative fact moved. It used to be the moment a request was time stamped at an official point of acceptance. It is now the moment the money is sitting in the scheme's account and is free to be used. Those two moments are the same only when the payment rail is fast, and they diverge by a full business day, sometimes more, whenever it is not.

The rule is a comparison of two clocks, not a deadline

Write the rule the way the industry's own operating notes write it, because the phrasing is exact and the paraphrases are not. The transaction request date and time, or the credit received date and time, whichever is later, is what determines NAV applicability.

That single sentence carries the whole mechanism. There are two events. Each has a timestamp. The later one is compared against the cut-off for the scheme category, and the answer to that comparison is the price.

Two consequences fall straight out of it. First, an early application cannot rescue slow money, because the comparison takes the later timestamp and the money is the later one. Second, fast money cannot rescue a late application either, because the comparison is symmetric. Prefunding a platform wallet on Monday and placing the order at 4:15 PM on Tuesday still gets Wednesday's NAV.

The phrase that does the heavy lifting is available for utilisation. It is a stronger condition than credited. A payment can be showing against a folio, visible in a platform's ledger and still not be money the scheme can deploy, because the funds are reconciled into the collection account of the fund and then released. The condition is about the fund's ability to invest the rupees, not about an investor's ability to see them.

The threshold that used to buy same-day pricing, and the gap it left open

Before February 2021 the rule split on amount. A purchase of two lakh rupees or more was already priced on realisation. A purchase below two lakh rupees was priced on the application timestamp, and the money could arrive whenever the payment instrument got round to it.

Read that carefully, because it is the reason the rule changed. Units were being created at a price struck on a day when the rupees behind them were still in somebody else's bank account. The scheme carried those units from the moment they were allotted. Whatever the portfolio did between the price being struck and the money landing was absorbed by the investors already holding the scheme.

The effect was largest exactly where it was least visible. A debt or short duration scheme accrues every calendar day. An application stamped on a Friday, funded by an instrument that cleared on the following Tuesday, was priced at Friday's close and picked up four days of the scheme's accrual on money that never left the applicant's account over the weekend. Nothing about that was illegal, and it did not require a large investor or any cleverness. It was simply how the threshold worked, and it was paid for by everybody else in the scheme.

The threshold is worth naming precisely because so much published material still describes it as live. An article that tells you a small investment gets same-day NAV on submission is not slightly out of date. It is describing a rule that was withdrawn more than five years ago, and every worked example in it prices on the wrong day.

The cut-off times did not go away, and they are not the same for every scheme

Realisation decides which day's money you have. The cut-off decides which side of that day you land on. Both tests have to be satisfied, and the second one is scheme specific.

Two rows in that table trip people up. The first is the liquid and overnight purchase, which gives the NAV of the day before the money arrives rather than the day of. That looks like a quirk and is not one: a liquid scheme accrues daily, so units have to exist from the start of the first day the money is actually working. Buying at the previous close is what makes that possible.

The second is the 7:00 PM online redemption cut-off for overnight schemes, which is new. It arrived through SEBI circular SEBI/HO/IMD/PoD2/P/CIR/2025/56 dated 22 April 2025 and took effect on 1 June 2025. The reason is worth knowing because it explains why only overnight schemes got it. A December 2023 requirement obliges stock brokers and clearing members to upstream client funds to clearing corporations at the end of each day, and one permitted form of upstreaming is a pledge over units of an overnight scheme. Those units have to be released and redeemed after the market closes, which is after 3:00 PM. The later cut-off exists to make that operationally possible. It does not extend to liquid schemes and it does not extend to purchases.

One further timing lives outside the regulation entirely. A platform that routes an order onward will impose its own internal deadline, earlier than the regulatory cut-off, because it must have the credit in hand before it passes the transaction on. Industry practice requires that transactions are only forwarded where the credit has already been realised. So the deadline that binds you is whichever is earliest among the platform's internal cut-off, the scheme's regulatory cut-off, and the moment your payment rail actually settles.

One instruction, two rails, two different days

Because the test is now about money rather than intent, the choice of payment rail is a pricing decision. It was not one before, and almost nobody treats it as one now.

Rails differ along three axes that matter here: whether they settle immediately or in scheduled batches, whether they operate outside banking hours and on non-working days, and whether the credit to the beneficiary is a separate event from the debit to the payer. Each axis can cost a day.

The practical reading is unglamorous. If you care which day you are priced on, pay through a rail that settles immediately and instruct well before the cut-off. If you do not care, the rail does not matter. What does not work is caring about the day and then paying through something that clears in batches, which is the combination almost every surprised investor has actually chosen.

A systematic instalment has three dates, and only one of them prices it

This is where the rule is met most often, and where the mental model fails most quietly. A recurring instalment involves an instalment date recorded on the registration, a debit date on which the mandate is presented and money leaves the investor's bank, and a credit date on which the money lands in the scheme's collection account. Allotment follows the third one.

The industry's own operating notes state it flatly: allotment of units is based on the date and time of receipt of funds into the collection account of the fund, and not on the debit date from the investor's bank account. Different payment modes have different clearing cycles, so a difference between the instalment date, the debit date and the final NAV date is normal rather than exceptional.

The last row is the one worth sitting with. Between the debit and the allotment the money has left the investor's account and has not yet bought anything. It is not earning in the bank and it is not exposed to the scheme. For a single instalment that is trivia. Repeated monthly for a decade it is a structural feature of the plan that nobody chose, and it is adjustable for free by moving the instalment date to a part of the week where the credit does not have to cross a weekend.

None of this makes an instalment worse. It makes the schedule a thing with a mechanism rather than a thing with a number. If you want to understand what that schedule is actually doing to your average cost, the companion piece on what a systematic investment plan actually is works through the averaging arithmetic itself.

Where the gap costs something: a large amount on a day that moves

For a routine instalment a one day slip is noise. The rule bites on a single large instruction placed on a day when the market is moving, because the whole amount is priced at one NAV and there is no averaging to absorb the difference.

Twenty five lakh rupees at an illustrative closing NAV of 214.6180 buys 11,648.604 units. The same amount priced one business day later at 221.3060 buys 11,296.576 units. The difference is 352.028 units, about 3.02 percent of the allotment, produced by a NAV movement of 3.12 percent across two adjacent business days.

Say plainly what that is and what it is not. It is not a loss, because the investor never held the units that were not allotted. It is a permanent difference in the size of the holding, arising from a day that a clearing cycle chose. Run the same slip across a day when the NAV fell instead and it allots more units, not fewer. The asymmetry is not in the direction, it is in the control: the investor decided the amount and the scheme, and something else decided the day.

The volatile day is also precisely when the slip is most likely, because a decision taken in reaction to a move is taken late. An instruction placed at 2:50 PM after watching the morning is inside the cut-off by ten minutes and stands almost no chance of having realised money behind it unless the rail is immediate. The behaviour that creates urgency and the behaviour that creates delay are the same behaviour.

Redemption did not change, and the asymmetry is deliberate

It is easy to assume that if purchases moved to a realisation test, redemptions must have too. They did not. A redemption is priced on the time a valid request is received, full stop.

The reason is structural rather than administrative. On a purchase there is an incoming payment whose arrival is genuinely uncertain, so the scheme cannot price until it knows. On a redemption there is nothing to wait for: the units already exist, the scheme already holds the assets, and the only fact in question is when the instruction arrived. There is no second clock to compare against.

What follows the redemption pricing is a separate timeline. Proceeds are transferred within three working days of the redemption, extended to five working days for schemes that hold at least eighty percent of total assets in permissible overseas investments. That payout window is not part of the NAV question, and conflating the two is how people convince themselves their redemption was priced late when it was priced on the day they asked.

The switch is the case almost everyone gets wrong

A switch looks like one instruction and is two. The switch-out is a redemption from the source scheme and is priced on the request timestamp. The switch-in is a purchase into the destination scheme and is therefore subject to the realisation test, funded by the settlement of the switch-out. The two legs are priced on different days whenever the source scheme has a settlement cycle longer than zero.

The equity row is the one with teeth. Since 1 February 2023 the payout cycle on an equity oriented redemption has been T plus 2, reduced from T plus 3, and the same cycle governs the money leaving an equity scheme on a switch. So a switch out of one equity scheme and into another is priced on T for the leg you are leaving and on T plus 2 for the leg you are entering. For two full business days the money is out of both schemes.

That is not a defect. It is the settlement cycle doing exactly what it is designed to do, and it is the price of moving money between two pools that both have to sell and buy. But an investor who believes a switch is an instantaneous transfer between funds has a model that will be wrong by two business days every time, and the last row shows the same mechanism stretching to five when the source scheme holds assets in another time zone.

It also explains a pattern that looks like bad luck. Switching out of a scheme because it has fallen, into one that has not, leaves the money unexposed for the two days in between. If you are switching because two schemes turned out to hold the same things, the diagnosis matters more than the timing, and the piece on measuring overlap between two schemes is the place to start.

Where this goes wrong in practice

Treating the acknowledgement as the answer. The confirmation you receive on submission carries the request timestamp. It is one of the two clocks, and on a slow rail it is the one that does not matter.

Paying at the last minute through a batch rail. The single most common version. An instruction at 2:45 PM is inside the 3:00 PM cut-off and the money is not, so the whole exercise buys nothing but a day of waiting.

Assuming the 3:00 PM cut-off is universal. It is 1:30 PM for purchases in liquid and overnight schemes, which is where large short term amounts are usually parked. A 2:30 PM instruction into a liquid scheme with money already in place is still an hour late.

Reading a platform deadline as the regulatory one. A routing platform sets its own earlier cut-off because it must hold the realised credit before it can pass the transaction on. The earliest deadline in the chain is the one that binds.

Expecting a switch to be same day on both legs. It is same day only within the same scheme. Out of an equity oriented scheme the switch-in is two business days behind the switch-out.

Reading a NAV date later than the instalment date as an error. It is the expected result whenever a mandate credit lands after the cut-off. If it happens every month, the instalment date is the thing to change, not the plan.

Using a page written before 2021. A great deal of live material still describes the two lakh threshold as current. Any page that says a small purchase gets same-day NAV on submission is describing a withdrawn rule, and its worked examples are all priced on the wrong day.

What the rule is actually for

Strip away the operational detail and the change did one thing: it made the price you pay depend on when your money joined the pool, rather than on when you said you intended to join it. That is not an anti-investor measure and it is not a convenience for fund houses. It is a statement about who bears the cost of a gap, and the answer it gives is that the person creating the gap bears it, rather than the people already in the scheme.

Everything practical follows from accepting that. The cut-off is a deadline for the money. The rail is part of the instruction. The instalment date is a variable you can set. The switch has two prices and two days. None of this requires prediction, and none of it can be optimised into an advantage, which is exactly why it is worth getting right: it is one of the very few parts of investing that is entirely under your control and entirely knowable in advance.

If you want to check where the rule bites in your own account, do it with a statement rather than from memory. Take three recent purchases, put the instruction timestamp, the bank debit date and the NAV date side by side, and see how often the third one is later than the first two. If the answer is always, you have found a fixed setting worth changing. If the answer is never, your rail is already fast and you can stop thinking about it. For the wider vocabulary this sits inside, the guide to what a mutual fund is in India covers the structure the NAV is computed from.