How to Use a Stock Screener: A Practical Beginner Guide
A stock screener is a filter you point at the entire market. You set numeric conditions, the tool discards every stock that fails even one of them, and you are left with a shortlist short enough to read properly in an evening.
A screener does not choose investments for you. It only removes the names that break your rules, which means the quality of your output is decided entirely by the quality of your inputs. A badly built screen either hands you 600 stocks or zero, and both outcomes feel like the software is broken when the real problem is the thresholds.
Below: the filter families that matter, how to build a first screen, a worked example against four conditions, and the mistakes that make screens useless.
What a screener does, and what it cannot do
Every screener runs on a database of numbers: price, traded volume, reported financials, valuation ratios, sometimes indicator values. You write conditions against those fields and the tool matches rows. That is the whole mechanism, which explains the limits.
A screener cannot tell you that a company’s profit jumped because it sold a factory. It cannot read a regulatory order, an auditor’s qualification, or a pledged promoter shareholding note. Reported numbers look backward, and screeners are only as fresh as their last data update.
Treat the output as a reading list, not a buy list.
The five filter families you will use most
You rarely need more than five or six conditions in total.
| Filter family | Typical fields | What it controls | Illustrative setting |
|---|---|---|---|
| Size and liquidity | Market cap, average daily traded value | Whether you can enter and exit without moving the price | Market cap above Rs 5,000 crore |
| Profitability | Return on equity, operating margin, net margin | How much profit the business earns on the capital it uses | ROE above 15% for three years |
| Balance sheet safety | Debt to equity, interest coverage, cash flow from operations | Survival odds in a bad year | Debt to equity below 0.5 |
| Growth | Sales growth, profit growth, five year compounded growth | Whether the business is expanding or drifting | Sales growth above 10% annually |
| Valuation | Price to earnings, price to book, dividend yield | What you are being asked to pay for all of the above | P/E below the sector median |
Liquidity deserves special respect. A stock trading Rs 40 lakh a day cannot absorb a Rs 5 lakh order without slippage, however good its ratios look. Traded quantity and spread are covered in our note on how to read a stock quote.
How do I build my first screen?
Start narrow, then loosen. Beginners do the opposite and drown.
- Write your idea in one plain sentence first. For example: “profitable mid sized companies that grow sales without borrowing heavily.”
- Translate each phrase into one numeric field. “Mid sized” becomes a market capitalisation band. “Without borrowing heavily” becomes a debt to equity ceiling.
- Add a liquidity floor before anything else. Average daily traded value above Rs 2 crore removes most untradeable names.
- Apply two quality conditions, not six. ROE and debt to equity carry a lot of information on their own.
- Run it, note the result count, then adjust one condition at a time so you can see which filter is doing the work.
- Export the survivors and read the annual report of the three that interest you most.
That fifth step is the one people skip. Change four thresholds at once, watch the count jump from 4 names to 90, and you have learned nothing about your own screen.
A worked example: does this company clear the filter?
Suppose your screen has four conditions: market cap above Rs 5,000 crore, ROE above 15%, debt to equity below 0.5, and P/E below 30.
Now take a hypothetical Company A:
- Share price: Rs 850
- Shares outstanding: 9 crore
- Net profit for the year: Rs 240 crore
- Shareholders’ equity: Rs 1,500 crore
- Total borrowings: Rs 600 crore
Work through it line by line.
Market cap = 9 crore shares multiplied by Rs 850 = Rs 7,650 crore. That clears the Rs 5,000 crore floor.
ROE = 240 divided by 1,500 = 0.16, so 16%. Clears the 15% condition, but only just.
Debt to equity = 600 divided by 1,500 = 0.40. Comfortably under 0.5.
Earnings per share = Rs 240 crore divided by 9 crore shares = Rs 26.67. P/E = 850 divided by 26.67 = 31.9. This fails the P/E ceiling of 30.
So Company A is rejected on one condition out of four, by roughly 6%. A rigid screen deletes the name. A thoughtful investor notes that it passes three quality tests and is only marginally expensive, and may widen the P/E limit to 35 to see what else appears. Both choices are defensible. Pretending the screener made the decision is not.
Why does my screen return 500 stocks or nothing at all?
Two failure modes, two different fixes.
Too many results
Your thresholds sit close to market averages, so most of the market passes. Tighten the condition most central to your idea rather than adding fields. Raising an ROE floor from 12% to 18% cuts a list harder than three extra filters.
Zero results
You have stacked conditions that rarely coexist. High growth plus a low P/E plus a high dividend yield plus zero debt describes a company that does not exist in most markets. Remove filters one at a time until names appear, and the last one you removed was doing the damage.
Mistakes beginners actually make
- Screening on a single year of data. One good year can come from a one-off item, so ask for three years of consistency.
- Comparing P/E across unrelated sectors. A bank, a cement maker and a software firm do not share a fair multiple.
- Skipping the statements because ratios felt like enough. Our walkthrough on how to read a balance sheet is the next step.
- Mixing valuation and momentum filters in one query, then being unable to explain why any name appeared.
- Re-running the same screen daily. Fundamentals move quarterly, not hourly.
Fundamental screens and chart based screens also answer different questions, which is why technical and fundamental analysis tend to be used at different stages of the same decision.
Frequently Asked Questions
Is a free stock screener good enough for a beginner in India?
Yes, for most purposes. Free tools cover market cap, standard ratios and growth rates, which is everything a first screen needs. Paid versions mainly add deeper history, custom formulas and backtesting. Learn to build clean screens on a free tool first, because paying for fields you cannot interpret changes nothing.
How many stocks should a good screen return?
Aim for 15 to 40 names. Below 10 usually means the conditions are so tight that you are fitting them to a few familiar companies. Above 100 means you will skim instead of read. What matters is whether you can genuinely study every survivor.
How often should I re-run my screen?
Once a quarter suits a fundamentals based screen, since that is when Indian companies report. Refresh a few days after the reporting deadline so the database has updated. Price based screens can run weekly. Daily re-running of a value screen generates activity rather than insight.
Can I screen for penny stocks safely?
You can, but add a liquidity floor and a minimum price condition, or you will surface names where one order moves the price several percent. Small companies also disclose less detail and often depend on a single customer or promoter. Position sizing matters more here than the screen.
Does a stock screener work for mutual funds too?
Separate tools exist for funds, filtering on expense ratio, assets under management, category, riskometer rating and rolling returns rather than P/E or debt to equity. The habit transfers even though the fields do not. Decide what you are optimising for before you filter.
Why do two screeners show different values for the same ratio?
Because they calculate differently. One may use standalone results and another consolidated, one trailing twelve month earnings and another the last full financial year, and treatment of exceptional items varies. Pick one source, learn its conventions, and verify anything decision critical against the filed statements.
Key Takeaways
- A screener removes stocks that fail your rules. It never ranks quality, so every shortlist still needs reading.
- Use five or six conditions across size, liquidity, profitability, balance sheet safety, growth and valuation.
- Put a liquidity floor, such as average daily traded value above Rs 2 crore, in every screen before you touch a ratio.
- Change one threshold at a time and record the result count, so you know which condition is filtering.
- A near miss like a P/E of 31.9 against a 30 ceiling is a judgement call, not a verdict.
- Screener data carries errors and lags. Confirm anything you act on against the company’s own filings.
