Money Clarity

How much to save every month, and where to put it

Most saving advice gives one number and one destination: put aside twenty percent, into a savings account. That fails for a reason that has nothing to do with discipline - money set aside for different jobs behaves differently, and a single pot gets emptied by whichever need shouts loudest. This is the four-bucket version, the order to fill them, and the arithmetic for how big each one should be.

Get Unyfy free See where your money goes

Last reviewed 2026-09-22

Start from committed cost, not from income

The technique

Income minus committed cost is the only number you can actually plan with

Percentage rules are applied to income because income is the number people know. But you cannot save out of income; you save out of what is left after obligations that do not move. Two people on Rs 60,000 with different EMI loads have entirely different capacity, and a rule that ignores that will prescribe the same amount to both.

Work out three figures, in this order, before deciding any savings target.

Committed cost is everything that leaves whether or not you pay attention: rent, EMIs, card minimums, insurance premiums, school fees, subscriptions, the money sent home. Variable cost is groceries, transport, eating out - things that flex. What remains is capacity.

The number worth watching every month is income minus committed cost. That is what you can actually direct.

A savings target set as a percentage of income ignores the EMI load, which is the thing that actually decides capacity
Committed cost creeping upward is the quiet failure: one new subscription and one EMI conversion, and the plan stops working without anyone deciding anything
If committed cost is above 70 percent of income, no savings plan will survive - the work is on the obligations, not on the saving
Review committed cost quarterly. It is the only line that changes without you noticing

Bucket one: the emergency fund

The technique

Size it on months of committed cost, not months of income

The common advice is three to six months of income. But in an actual emergency - a job loss, a medical event - variable spending collapses on its own. What must keep being paid is committed cost. Sizing on income overstates the target by the amount you would stop spending anyway, which makes the goal feel unreachable and is a common reason people never start.

The point of this fund is not the interest it earns. It is the interest it prevents.

The same Rs 40,000 emergency, two ways

On a credit card at 3.5%/month plus GST, cleared over 12 months
monthly instalment: Rs 4,139
total repaid: Rs 49,672
cost of the emergency: Rs 9,672
From a fund earning 3.5% a year
interest given up over the year: Rs 1,400
Difference: Rs 8,272

A Rs 40,000 buffer pays for itself the first time it is used. It only has to be needed once every seven years to beat the interest it forgoes by sitting there.

Income	Committed cost	3 months	6 months
Rs 35,000	Rs 27,000	Rs 81,000	Rs 1,62,000
Rs 60,000	Rs 38,000	Rs 1,14,000	Rs 2,28,000
Rs 1,00,000	Rs 62,000	Rs 1,86,000	Rs 3,72,000

Three months is the floor. Six is right for a single income, variable pay, or self-employment.

Keep it somewhere boring and reachable in a day - a sweep account or a liquid fund, not equity and not locked away
Keep it in a separate account with no card attached. The friction is the feature
Do not chase yield here. The return on this money is the borrowing it prevents, not the interest it earns
Rebuild it before resuming anything else after you use it. That is what makes it a fund rather than a one-off

Bucket two: the expensive months

The technique

Sinking funds - annualise the irregular, divide by twelve

Plans almost never break in a normal month. They break in the month with a wedding, a festival, an insurance renewal or a school fee - costs that are entirely predictable a year ahead and get treated as emergencies when they land. A sinking fund converts a shock into a monthly line.

Write down every irregular expense of the last twelve months - not groceries, the ones that arrive once or twice a year and always feel like bad luck. Total them and divide by twelve. That figure has been part of your cost of living all along; it simply has not been budgeted.

Use last year's actuals, not next year's estimate. Your own history is the only reliable forecast
Include the ones you would rather forget: gifts, repairs, the trip home you always end up taking
Hold it separately from the emergency fund. Mixing them means a predictable cost quietly eats the buffer for unpredictable ones
When a sinking fund is properly stocked, a wedding season stops being a reason to use a credit card

Bucket three and four, and the order that matters

The technique

Fill in sequence, because the buckets have different returns

The order is not preference, it is arithmetic. Money in an investment earning 11 percent while a card balance runs at 42 percent is losing 31 percent a year with extra steps. Every rupee has exactly one best home at any moment, and it changes as the buckets fill.

Bucket three is high-interest debt. Bucket four is long-term investing - retirement, a house, education. The sequence below is what to do with each additional rupee of capacity.

1. One month of committed cost in the emergency fund. A small buffer stops the next minor shock becoming debt
2. Any employer match on a retirement contribution. It is the only guaranteed hundred percent return available to anyone
3. Every balance above roughly 15 percent - cards first, then consumer durable, then personal loans. Nothing else you can do with money beats this
4. The emergency fund up to three months, then six if your income is variable or single
5. The sinking fund to its monthly figure
6. Long-term investing, and only now. Systematic, automatic, and left alone
7. Low-rate debt - a home loan at 8.5 percent is usually better kept than prepaid once everything above is done

Making it happen without relying on willpower

The technique

Automate on the salary date, and escalate with each raise

Thaler and Benartzi tested this directly. Employees offered their Save More Tomorrow plan - contributions automatic, and rising with each pay raise rather than with each decision - took it up at 78 percent, and 98 percent were still in it two raises later. Average savings rates went from 3.5 percent to 11.6 percent across three raises over 28 months. Nobody was persuaded to be more disciplined; the decision was simply removed.

Saving a fixed amount is easy. Deciding to save it twelve times a year is hard, and it is the deciding that fails - usually around the 20th, when the balance looks thin.

Set a standing instruction for the day after your salary credits, when the account is at its fullest. Not the 25th, when it is at its emptiest and the transfer bounces or gets cancelled.

Salary date plus one, not a round number mid-month
Start at an amount that survives a bad month. Rs 500 that runs for two years beats Rs 5,000 that stops in month three
Raise it on the day your income rises, not at the start of the next year. That is the whole mechanism of Save More Tomorrow in miniature
Separate accounts per bucket. One pot with four jobs gets emptied by whichever job shouts loudest
Automate the transfer, not the decision to make the transfer

What to be careful of

The failures here are predictable and mostly involve a product being sold as a plan.

An insurance policy is not a savings plan. Endowment and money-back products bundle poor returns with thin cover - buy term cover for protection and invest separately
Do not lock the emergency fund into anything with a penalty or a notice period. A buffer you cannot reach in a day is not a buffer
Guaranteed high returns do not exist. Anything promising them is either mispriced risk or a fraud
Do not invest while carrying a balance above about 15 percent. It feels like progress and it is a net loss
A plan that needs you to remember it every month will fail in month three. If it is not automatic, it is a wish

How Unyfy helps size savings from committed cost

This page builds every bucket on one figure, income minus committed cost, and the hard part is getting committed cost right rather than guessing it. The app reads your bank and card transaction emails and, on Android, your bank's transactional SMS, with no manual entry, and from the payments that repeat on a cycle, on Pro, it predicts what the coming month is already committed to.

On Pro, the Fixed Expenses screen shows next month's committed outflows, the EMIs, SIPs, rent, bills, subscriptions and card bill, with what is paid and what is left so far this month, and a Subscriptions list with each recurring subscription, its amount and whether it is due or paid. The gap between salary and commitments becomes a number you read rather than estimate. That gap is what sizes the emergency fund in months and what a standing instruction on salary day can safely carry.

Set up the transfer to a separate account, and the step-up on each raise, with your bank. It never asks for your bank password or UPI PIN, and every payment is one you authorise.

Install Unyfy on Android, or use the web app at app.unyfy.co.in on an iPhone.

Common questions

How much should I save every month in India?

Start from income minus committed cost rather than from a percentage of income, because your EMI load decides your capacity and a percentage rule ignores it. Then fill in order: one month of committed cost as a buffer, any employer retirement match, every balance above roughly 15 percent interest, the emergency fund to three months of committed cost (six if your income is variable or single), the sinking fund for irregular expenses, and only then long-term investing. The amount matters less than the order and the automation.

How big should my emergency fund be?

Three to six months of committed cost, not of income. In a real emergency variable spending collapses on its own; what must keep being paid is rent, EMIs, insurance and fees. On Rs 60,000 income with Rs 38,000 of committed cost, three months is Rs 1,14,000 and six is Rs 2,28,000. Sizing on income would have suggested Rs 1,80,000 and Rs 3,60,000 - a target inflated by spending you would have stopped anyway.

Should I invest or pay off my loan first?

Clear anything above roughly 15 percent before investing. A credit card at 42 percent against an investment returning 11 percent is a net loss of about 31 percent a year, whatever the investment statement says. Below roughly 9 percent - most home loans - investing usually wins once your emergency fund is in place. Between the two, it depends on tenure and tax, and either choice is defensible.

Is an emergency fund worth it when savings accounts pay so little?

Yes, because the return is the borrowing it prevents rather than the interest it earns. A Rs 40,000 emergency put on a credit card at 3.5 percent a month plus GST costs Rs 9,672 over a year. The same Rs 40,000 sitting in an account at 3.5 percent a year gives up Rs 1,400 of interest. The fund is ahead by Rs 8,272 the first time it is used, and only needs to be used once every seven years to break even against the interest it forgoes.

What is a sinking fund and how do I calculate mine?

It is money set aside monthly for expenses that arrive once or twice a year - festivals, weddings, insurance renewals, school fees, annual repairs. List what you actually spent on these over the last twelve months, total it, and divide by twelve. That figure has always been part of your cost of living; the only change is budgeting for it. Keep it separate from the emergency fund, or a predictable cost will quietly consume the buffer meant for unpredictable ones.

Saving is not one number and one account. It is four funds with different jobs, filled in an order set by arithmetic rather than preference - a small buffer, then any employer match, then every balance above 15 percent, then the emergency fund to three or six months of committed cost, then the sinking fund, then long-term investing. Size the buffer on what must keep being paid rather than on income, because a Rs 40,000 emergency costs Rs 9,672 on a card and Rs 1,400 from a fund. And automate it on the day your salary lands, because the deciding is what fails, not the saving. Unyfy computes your committed cost and your real surplus from the accounts themselves, finds the irregular spending you have not budgeted for, and tells you which bucket the next rupee belongs in.

Get Unyfy free

Banks and NBFCs we compare

Unyfy compares offers from these lenders and earns a commission if you take one. The comparison is shown first, and it can tell you not to switch.

Axis Bank

How much to save every month, and where to put it

Start from committed cost, not from income

Bucket one: the emergency fund

Bucket two: the expensive months

Bucket three and four, and the order that matters

Making it happen without relying on willpower

What to be careful of

How Unyfy helps size savings from committed cost

Common questions

How much should I save every month in India?

How big should my emergency fund be?

Should I invest or pay off my loan first?

Is an emergency fund worth it when savings accounts pay so little?

What is a sinking fund and how do I calculate mine?

Related topics

Emergency fund: size it on what cannot stop, not on what you spend

How much of your income can safely go to debt?

Where does my salary go? Follow the waterfall by date

How to save more money in India without earning more

What is Unyfy?

Banks and NBFCs we compare