Most portfolios are built to be right. They hold what the investor expects to do well, in the amounts conviction suggests, and they are reviewed when the news demands it. A resilient portfolio is built on a different assumption: that some of its views will be wrong, that the timing of markets is unknowable, and that the investor holding it is a human being who will be tempted to abandon it at the worst moment. This guide sets out the sequence we use to build one — allocate, size, systematise, rebalance — and links to the deeper articles on each step.
What “resilient” actually means
Resilient does not mean safe, and it does not mean high-returning. It means the portfolio can absorb a bad outcome without forcing a bad decision.
Three tests capture it. First, no single position or asset class can impair the plan on its own. Second, the portfolio can be held through a deep drawdown without the investor being pushed into selling — whether by cash needs or by nerves. Third, the rules for what to do next exist before the event, not after it.
Notice that none of these tests mention returns. That is deliberate. Return is what the market gives you over time for bearing risk sensibly; resilience is what keeps you in the market long enough to collect it. Most portfolio failures we see are not selection failures. They are structural failures: too much of one thing, held with no rule for when to trim, sold in a panic. The steps below address the structure.
Step 1: Start with goals and horizon, not products
A portfolio is a tool for funding a life, so the design has to begin with the life. Before any asset is chosen, write down what the money is for, when it will be needed, and what a shortfall would actually cost.
Horizon does most of the work. Capital needed in two years cannot tolerate much volatility, however attractive the opportunity. Capital not needed for fifteen years can, and probably should, because the cost of holding too much in cash over that span is real, even if it never appears on a statement as a loss.
Separate the goals rather than pooling them. A house deposit due in 2029, a child’s education in 2035 and retirement in 2050 are three different problems with three different horizons, and they should not share one allocation.
Step 2: Asset allocation is the decision that matters most
Once the goals are clear, the single most important decision is how capital is split across asset classes — equity, debt, gold, cash and, where appropriate, real assets. Everything downstream is a refinement of this choice.
The evidence for its weight is old and well tested. Brinson, Hood and Beebower’s 1986 study of US pension plans in the Financial Analysts Journal found that investment policy — the long-run asset mix — explained over 90 percent of the variation in quarterly returns, dwarfing market timing and security selection. The finding is often misquoted as “allocation explains 90 percent of returns”; it does not. It explains most of why a portfolio’s returns move the way they do, which is exactly what resilience is about.
We covered the reasoning in why asset allocation is the only free lunch in investing: asset classes respond differently to the same conditions, so the mix, not the picks, decides how the portfolio behaves under stress.
What is a good asset allocation in India?
There is no universal answer, and anyone who offers one without asking about your goals is selling a product, not advice. But the shape of a sensible answer is consistent.
Equity carries the growth burden for long horizons. Indian equity is volatile, and international equity adds a second economy and a second currency, which matters when the rupee moves. Debt — government securities, high-quality corporate bonds, and the funds that hold them — dampens volatility and funds nearer-term goals. Gold has historically behaved differently from both in Indian conditions, and a modest, deliberate allocation is common. Cash is for liquidity, not return.
The proportions come from Step 1. A longer horizon and a higher capacity for loss push the equity share up; a nearer goal or a thinner safety net pushes it down. The right allocation is the most aggressive one you can hold through a bad year without breaking the plan — not the most aggressive one you can imagine in a good year.
Step 3: Diversification that actually diversifies
Diversification is measured by correlation, not by count. A portfolio of six large-cap Indian equity funds is one position wearing six names; they will rise and fall together because they own largely the same companies.
Real diversification means holding things that do not depend on the same outcome. Domestic equity and international equity depend on different economies. Equity and government debt depend on different phases of the interest-rate cycle. Gold depends on a different set of fears altogether. When one of these is under pressure, another is often providing ballast, and that offset is what lets the whole survive.
The test to apply is simple. For each holding, ask what conditions would make it fall, then check how many of the others would fall in the same conditions. If most of the portfolio shares one failure mode — a sharp rise in rates, a domestic slowdown, a currency shock — it is not diversified, whatever the holding count says.
How many funds should a portfolio have?
Fewer than most investors own. The number is a consequence of the allocation, not a goal in itself.
A useful discipline is to start from the asset classes the plan needs and give each one a small number of vehicles that cover it cleanly. Domestic equity might be one broad fund, or a broad fund plus a deliberate mid- or small-cap tilt. International equity, one or two. Debt, one or two matched to the horizon of the goals they fund. Gold, one. That is often six to ten holdings for a complete plan.
Beyond that, additional funds tend to add overlap rather than coverage. Each new fund also adds a line to track, a rebalancing decision, a tax event on exit and a reason to tinker. Portfolios with twenty-five funds are not more diversified than portfolios with eight; they are usually less understood. When you cannot say in one sentence what job a holding does that no other holding does, it is a candidate for removal.
Step 4: Position sizing
Allocation decides how much goes to each asset class. Sizing decides how much goes to each holding within it, and this is where most avoidable damage originates.
The instinct is to size on conviction: the more sure you are, the more you buy. But conviction peaks when a story is most compelling, which is not the same as when it is most likely to be right. The more reliable input is consequence: if this is wrong, what does the portfolio look like afterwards? A holding whose failure would impair the plan deserves a smaller size than one whose failure would be an annoyance, whatever the confidence attached to each.
Our position sizing framework sets out three questions to ask before any size is chosen — what being wrong costs, what else moves with it, and whether you would hold it at a loss. Size is the risk control that operates before the first adverse move. Stop-losses and hedges operate after.
Step 5: Write the rules down
A plan that lives in your head is not a plan. It is a mood, and it will change with the market.
The rules that matter are few, and they should be written before they are needed: the target allocation and the bands around it; the maximum size of any single holding; what triggers a review; and what would have to be true for the allocation itself to change. The last one is the most important, because it separates a legitimate change in circumstances — a new goal, a windfall, a shortened horizon — from a reaction to a headline.
We have argued that the best investors build systems, not predictions, and this is the practical form of that idea for an individual. A written system does not need to be elaborate. It needs to be specific enough that a decision under pressure can be checked against it, and stable enough that it is not rewritten every time the market moves. Conviction is an input to a system, not a substitute for one.
Step 6: Rebalancing discipline
A portfolio built at a deliberate allocation does not stay there. After a strong equity run, the equity share rises and the portfolio becomes more aggressive than anyone chose — precisely when the investor feels most comfortable with risk. After a fall, the reverse.
Rebalancing restores the allocation you set when you were thinking clearly. It asks you to trim what has done well and add to what has not, which feels wrong every single time. That discomfort is why it needs to be a rule rather than a judgement.
It also helps to be honest about what rebalancing does. As we explained in rebalancing is a risk decision, not a return one, whether it adds or subtracts return in any given period is unknowable in advance. What it reliably does is keep the portfolio’s risk close to what the investor can actually tolerate — and tolerance is what determines whether someone stays invested through the difficult stretch.
How often should you rebalance?
Less often than the urge suggests, and always by a rule set in advance. Two approaches work; the choice between them matters less than having one.
Calendar rebalancing reviews the portfolio on a fixed schedule — annually or half-yearly is common — and resets it to target regardless of how markets feel that day. It is simple, predictable and easy to combine with a tax-year review.
Threshold rebalancing sets a band around each target and acts only when an allocation drifts outside it. A 60 percent equity target with a five-point band means nothing happens between 55 and 65 percent; a move beyond either edge triggers a reset. This responds to what the portfolio has actually done rather than to the date, and it leaves a well-behaved portfolio alone.
Many investors combine them: check on a schedule, act only if a band is breached. Whichever you pick, rebalancing too frequently creates costs and tax events for little risk benefit, and rebalancing “when it feels right” is not rebalancing at all.
Step 7: Costs, taxes and behaviour in an Indian context
Three things quietly decide how much of a portfolio’s return the investor actually keeps.
Costs. Every mutual fund charges a total expense ratio. SEBI caps what a scheme may charge under Regulation 52 of the Mutual Fund Regulations, with limits that fall as a fund’s assets grow, and the framework was revised again from April 2026. Within those caps, the gap between a low-cost and a high-cost fund covering the same market is a drag that compounds for as long as you hold it. Direct plans of the same scheme carry lower expenses than regular plans because they exclude distributor commission.
Taxes. For equity-oriented funds and listed shares, gains on units held more than twelve months are long-term and, as of assessment year 2026-27, taxed at 12.5 percent on aggregate gains above ₹1.25 lakh in a financial year; gains on shorter holdings are taxed at 20 percent, plus applicable surcharge and cess (Bajaj Finserv AMC, 2026). Rates change with budgets; the durable point is that every rebalance and every fund switch is a taxable event, so fewer, larger, deliberate trades beat frequent small ones. Debt funds follow different rules and should be checked separately.
Behaviour. A systematic investment plan is often described as a return strategy. It is better understood as a behavioural device: it automates the decision to invest, removes the question of timing from each month, and keeps buying when the investor would otherwise have paused. For most salaried investors that is worth more than any view on valuations.
A worked illustration: ₹50 lakh with rebalancing bands
The following is for illustration only. The allocation is not a recommendation, and the price moves are chosen to make the arithmetic clear, not as forecasts.
Suppose an investor with a long horizon sets a ₹50 lakh portfolio at 60 percent equity, 30 percent debt and 10 percent gold — ₹30 lakh, ₹15 lakh and ₹5 lakh — with a five-point band around each target, reviewed twice a year.
Imagine a strong year in which the equity holdings rise 40 percent to ₹42 lakh, debt gains 6 percent to about ₹15.9 lakh and gold is flat at ₹5 lakh. The portfolio is now worth about ₹62.9 lakh, and equity has drifted to roughly 67 percent — outside the 65 percent upper band. The rule says act: sell around ₹4 lakh of equity and move roughly ₹3 lakh into debt and ₹1 lakh into gold, bringing each back near target. Nothing about the decision required an opinion on whether equities would keep rising.
Now imagine the reverse: equity falls 30 percent to ₹21 lakh while debt and gold hold. Equity is now under 50 percent of a portfolio worth about ₹42 lakh, below the 55 percent lower band, and the same rule says buy equity with proceeds from debt. That purchase will feel reckless at the time. It is simply the plan.
The illustration ignores taxes and transaction costs, both of which would apply to the sale, which is one reason bands are set wide enough that the rule fires rarely.
Common mistakes that make a portfolio fragile
Most fragility comes from a short list of habits, all of which feel reasonable while they are being formed.
- Letting winners run without a rule. The largest holding is usually the one that has done best, which means the portfolio’s risk is concentrated in the position the investor is least willing to trim.
- Confusing fund count with diversification. Many funds holding the same large-cap names is one bet, many times over.
- Holding no debt because “rates are low.” Debt is there for its behaviour in a drawdown, not for its yield in a boom.
- Building the portfolio before the emergency fund. A forced sale in a downturn is the single most expensive transaction an investor can make.
- Changing the allocation after a bad year. A plan that only survives good years was not a plan.
- Ignoring the tax cost of tinkering. Each switch resets the holding period and may crystallise a gain.
- Chasing last year’s category. Sector and thematic funds are bought at the top of their story far more often than at the bottom.
None of these are errors of intelligence. They are errors of structure, and structure is fixable.
When should the plan itself change?
A written plan is not a permanent one, but the reasons to change it should come from your life, not from the market.
Legitimate triggers are specific: a goal has moved closer or been met, income has changed materially, a dependant has been added, a large inheritance or liquidity event has arrived, or you have lived through a drawdown and learned that your real tolerance for loss is lower than you assumed. Each of these changes the inputs to Step 1, so the allocation that follows from them should be revisited.
Illegitimate triggers are just as specific: a fund manager’s new outlook, a strong or weak quarter, a headline about the economy, a friend’s returns. If the reason for a change could not have been anticipated in the written rules and does not alter your goals or horizon, it is a reaction, and the plan exists precisely to outlast reactions.
Go deeper
Each step above has its own article. Read them in this order for the full sequence.
- Asset allocation: the only free lunch in investing — why the mix outweighs the picks.
- Position sizing: the decision most investors skip — three questions to ask before any size is chosen.
- Why the best investors build systems, not predictions — what a written process looks like and why it compounds.
- Rebalancing is a risk decision, not a return one — drift, bands and making the discipline mechanical.
The order matters less than the habit. A portfolio built this way will not be the best-performing one in any given year. It is designed to still be there, intact and understood, in the year that matters.
