How to build a stock portfolio from scratch

What are the actual steps to build your first stock portfolio in India?

Build a stock portfolio from scratch in four steps: keep six months of expenses in cash first, decide a fixed monthly amount, then buy 12-20 businesses across five or six unrelated sectors, each 4-8% of the portfolio. Screen for quality before price. Add every month for years, and sell only when the business breaks.

Key Takeaways

• Emergency fund and high-interest loans come before your first share purchase.
• A steady monthly amount builds better portfolios than occasional bonus money.
• 12-20 stocks is the practical range: enough spread, few enough to actually follow.
• Only 22.0% of Indian companies with full fundamentals clear a basic quality bar (September 2026).
• Cap any single stock near 10% and any single sector near 25%.
• Sell when the business changes, not when the price moves.

How much money do I need to start a stock portfolio?

Less than most people assume. On the NSE and BSE you can buy a single share, so the technical minimum is the price of one share plus brokerage — often under ₹1,000. The real limit is not the market, it is your own balance sheet. Before the first buy, park six months of household expenses in a savings account or liquid fund, clear any credit card dues or personal loan charging double-digit interest, and make sure nothing you invest is needed within the next five years. Money that has a deadline does not belong in equities.

Then fix a monthly number you can pay without arguing with yourself — ₹5,000, ₹15,000, ₹50,000, whatever survives a bad month. A portfolio built from a steady monthly amount behaves completely differently from one built on bonus money and sudden enthusiasm. The size of the amount matters far less than the fact that it keeps arriving, month after month, for years. If you are currently sitting on losses from a few impulsive buys, this is the step that was missing.

Indian listed companies clearing each screen, out of 1,518 with full-year fundamentals (September 2026)

Screen	Companies passing	Share of 1,518
ROE above 15%	458	30.2%
ROCE above 15%	464	30.6%
Debt-to-equity below 1	1,282	84.5%
PE between 0 and 40	988	65.1%
All three quality screens together	334	22.0%
Quality screens plus PE under 40	241	15.9%

How many stocks should a beginner's portfolio hold?

Twelve to twenty for most people. Below about eight, one accounting scandal or one collapsed sector takes a bite you will still remember in a decade. Above about twenty-five, you cannot honestly follow the businesses — you will not read the annual reports, you will not notice borrowings creeping up, and you end up owning something close to an index while doing the work of a stock picker.

Start smaller than your target. Four or five businesses in the first six months is perfectly fine. You are learning your own temperament as much as you are buying assets, and how you behave during a 30% fall in a name you researched carefully is information no book gives you. Add a new name only when you can explain, in two plain sentences, what the company sells and why it earns more than it costs to run.

Resist the urge to hold a stock in every theme that is being discussed. A portfolio is not a collection; it is a short list of businesses you chose on purpose.

How do I choose which stocks go into the portfolio?

Quality first, price second. Three numbers do most of the filtering: return on equity above 15%, return on capital employed above 15%, and debt-to-equity below 1. The first two say the business earns a decent return on the money sitting inside it. The third says it is not running on borrowed time. Only after a company clears those do you ask what you are being charged for it.

That bar is narrower than it sounds. In our universe of 1,847 listed Indian companies as of September 2026, 1,518 had full-year fundamentals — and only 22.0% of those cleared all three tests at the same time. Add a valuation check of PE under 40 and 15.9% survive, with the median PE across the universe at 24.0. So roughly one listed company in six is both a reasonable business and not obviously expensive. The useful part is that this still leaves 241 names, which is more than enough to fill a twenty-stock portfolio ten times over. Most of your job is saying no.

One honest exception: a flat debt-to-equity rule punishes banks and NBFCs unfairly, because borrowing is literally their business model. For lenders, look at capital adequacy, net interest margin, and how bad loans have moved over several years instead of the debt ratio.

How much should I invest in each stock?

Roughly equal weights, with a ceiling. For a sixteen-stock target, 5-7% in each works cleanly. Then set a hard rule: no name gets fresh money once it crosses 10% of the portfolio, however convinced you feel. Convinced is precisely the emotion people report just before a position teaches them something expensive.

Illustrative figures only: on a ₹4,00,000 portfolio with a 6% target weight, a full position is ₹24,000. If you invest ₹20,000 a month, that works out to roughly one new position every five weeks, or top-ups into names you already own. Winners will drift above their target weight on their own over time, and that is fine — letting a good business grow into a large holding is different from you manually pushing one stock to 25% because it has been rising.

Keep position sizes boring. Nobody has ever been ruined by equal weighting; plenty of portfolios have been ruined by one oversized conviction bet.

Which sectors should a first portfolio cover?

Five or six sectors that do not fail together. Four private banks is not diversification — it is one bet on credit growth wearing four different shirts. Spread across groups whose fortunes move for different reasons: lenders, IT services, consumer goods, pharmaceuticals, industrials and capital goods, autos, speciality chemicals. When one of them has a terrible year, the others should have no particular reason to join in.

Cap any single sector near 25% of the portfolio, and remember that your salary is also a sector bet. If you work in IT and hold employer stock in IT, a portfolio that is 40% IT services means one industry downturn hits your income, your options and your savings in the same quarter.

Keep the core in large, established businesses for the first two years at least. Small caps are where both the exciting returns and the ugly exit problems live — on a bad day you may find no buyer at a sensible price. They deserve a slice of the portfolio, not its foundation.

Should I invest a lump sum or buy every month?

Every month, in almost every case. Not because staggering magically earns more, but because it removes the worst decision available to a new investor: committing your entire savings during a month that happened to feel comfortable. Several brokers now allow a monthly order on individual stocks; a recurring calendar reminder does the same job for free.

If a large one-time amount has landed — a bonus, a property sale, a retirement payout — spread it over six to twelve months rather than one afternoon. Illustrative: ₹12,00,000 deployed as ₹1,00,000 a month across a year. Sometimes you will pay more than if you had invested everything on day one. You will also never be the person who put it all in three weeks before a 20% correction and then stopped investing entirely for five years. That second outcome is the one that actually destroys wealth.

What accounts and tax rules do I need to sort out in India?

A demat and trading account with a SEBI-registered broker, KYC completed with PAN and Aadhaar, a bank mandate, and a nominee. Add the nominee on the same day you open the account. Families spend years chasing unclaimed shares because one field was left blank.

On tax, as the rules currently stand, listed equity held for more than twelve months is long-term and taxed at 12.5% on gains above ₹1.25 lakh in a financial year; sold earlier, gains are short-term and taxed at 20%. Dividends are added to your income and taxed at your slab rate. Securities transaction tax is deducted on every trade, and brokerage and exchange charges sit on top. The practical lesson for a new portfolio is unglamorous: churning is taxed harder and costs more than sitting still.

Keep a one-page record for every purchase — date, price, and two lines on why you bought it. It is the cheapest portfolio tool in existence. When a holding drops 25%, that note is the only thing standing between a reasoned decision and a panicked one.

When should I sell a stock from my portfolio?

Sell when the reason you bought has stopped being true. Concretely: margins shrinking for several quarters with no explanation, debt rising while profits do not, the promoter group quietly reducing its stake, auditors resigning, or the company wandering into businesses it has no track record of running. Those are changes in the business, and they justify an exit.

Trim separately for housekeeping — a position that has outgrown your ceiling, or money genuinely needed for a life goal. What does not qualify: a falling price on its own, a flat year, or a neighbour's stock doing better. If you cannot name the specific thing that changed inside the company, you are reacting to a quote screen, not to information.

What mistakes wreck a new portfolio in its first year?

Four, in order of damage. Acting on a tip from WhatsApp, YouTube or a Telegram channel without checking a single number. Buying twenty-rupee shares because twenty rupees feels cheap — price per share tells you nothing about value. Averaging down into a company whose fundamentals are visibly deteriorating, which is how a small loss becomes a large one. And checking the portfolio several times a day, which turns ordinary volatility into anxiety and anxiety into bad trades.

There is a fifth: quitting in month seven. A portfolio of solid businesses does almost nothing interesting in its first two years. The compounding everyone talks about shows up in years five to fifteen, which is exactly why so few people ever experience it. Boredom is not a sign the method is failing.

Click Here – See BossInvestor's Data-Driven Stock Screens

---

Conclusion

Building a portfolio from scratch is mostly administrative work done in the right order: an emergency fund, a fixed monthly amount, a quality filter, 12-20 names across unrelated sectors, equal-ish position sizes, and written reasons you can check later. The screening arithmetic carries most of the weight — as of September 2026, just 15.9% of companies with full fundamentals were both decent businesses and sensibly priced, so saying no is the main activity. The method above is yours to use for free; which specific names clear that bar at today's prices is what our research covers inside the app once KYC is done.

BossInvestor is a SEBI Registered Research Analyst (INH000024143). This article is for educational and informational purposes only. It explains a method and is not investment advice, nor a recommendation to buy, sell or hold any security. Market-wide figures are computed from our universe as of September 2026; past metrics do not guarantee future results. Always do your own research or consult a registered adviser before investing.

---

Frequently Asked Questions

How many stocks is too many for a retail investor in India?

Past roughly twenty-five, the portfolio stops being yours in any meaningful sense. You cannot read that many annual reports, follow that many quarterly results, or notice when borrowings start climbing. You end up with index-like returns while carrying the workload and tax friction of active picking. If you want exposure to more than twenty-five companies, an index fund does it cheaper and better. Keep a direct portfolio to names you can genuinely track and explain.

Can I really build a portfolio with ₹5,000 a month?

Yes. Since single shares trade on the NSE and BSE, ₹5,000 a month is enough to begin, though it changes your sequencing. Rather than buying twelve stocks thinly, buy one good position at a time and build to twelve over a year or two. Many brokers charge low or zero delivery brokerage, so costs will not eat you alive. What matters is that the ₹5,000 arrives every month without fail, including in months when markets look frightening.

Should my first portfolio include mutual funds as well as stocks?

For most beginners, yes. An index fund or two can form the base while your direct stock portfolio grows alongside it. That way a mistake in your first few picks does not decide your entire outcome, and you still learn by owning businesses directly. A common structure is a core of index funds with a satellite of 12-20 individually chosen stocks. As your research process proves itself over several years, you can shift the weighting.

Why do so few Indian companies pass a basic quality screen?

Because earning more than 15% on both equity and capital employed, while keeping debt below equity, is genuinely difficult over a full year. Of the 1,518 companies with full-year fundamentals in our universe as of September 2026, only 22.0% managed all three at once, and 15.9% also traded under a PE of 40. The listed market contains thousands of businesses; only a minority are consistently good ones. The screen is not harsh — reality is.

How long before a new portfolio starts showing results?

Plan in years, not quarters. The first one or two years are mostly about building positions, paying tuition through small mistakes, and learning how you behave in a falling market. Returns in that window are noise and tell you almost nothing about your method. Judge the process instead: are you investing every month, are your holdings still clearing their quality tests, are your sizes under control? If those three are true, the arithmetic takes care of the rest.