Stock Screeners Explained: How to Use One Properly in India

A stock screener is a filter over the market. You set conditions — profit growth above this, debt below that, market capitalisation in this range — and it returns the companies that currently satisfy them. That is the whole mechanism, and seeing it plainly is what separates investors who get value from a screener from those who generate long lists and act on none.

India has capable screening tools, including Screener.in and Tickertape, and most brokers offer some form of screen. The tool matters far less than the screen you build with it. What follows is how to build useful screens, and the mistakes that make one worthless.

What a stock screener actually is

Every listed company reports numbers, and a screener is a query interface on top of that data. When you screen you are asking a database a question about the present: which companies currently meet these conditions? That is powerful — it turns thousands of names into twenty you can read about this weekend — but narrow. It does not know why a company is cheap, whether last year's growth is repeatable, or whether the accounting is conservative. Everything interesting happens afterwards.

The five filter families that matter

Screeners offer hundreds of fields, which is why most people use either one filter or twenty. Almost everything useful falls into five families, and a good screen draws from several so each covers the blind spots of the others.

1. Valuation: what am I paying?

Price-to-earnings, price-to-book, EV/EBITDA, dividend yield. These control the price you pay and are the most misused family in screening. A low P/E is not a bargain, it is a question — companies trade cheap for reasons: earnings at a cyclical peak, structural decline, governance concerns, a one-off gain inflating last year's profit. Two habits help. Use valuation as a ceiling rather than a target, excluding what is plainly expensive instead of hunting the cheapest. And compare within an industry, because a normal multiple for software is an extraordinary one for a cement plant.

2. Growth: is the business getting bigger?

Revenue and profit growth over three and five years, and the trend inside it. Long windows matter more than the latest quarter, because one good quarter is noise and easy to produce off a weak base. Require growth across more than one period so you find consistency rather than a spike, and be wary when profit growth runs far ahead of revenue growth for years: that is margin expansion, welcome but not indefinitely repeatable.

3. Quality: is the growth worth having?

Return on equity, return on capital employed, debt-to-equity, interest coverage, and operating cash flow against reported profit. This is the family beginners skip and the one that removes the most bad outcomes. Growth funded by debt at low returns on capital destroys value while looking impressive in a revenue line. And a persistent gap between reported profit and operating cash flow is the most useful red flag a screener can hand you.

4. Momentum: is the market agreeing yet?

Price relative to its 200-day moving average, distance from the 52-week high, relative strength against the index. Momentum is uncomfortable for value-minded investors and hard to ignore honestly. Used sensibly it confirms rather than justifies: a cheap, high-quality company falling every week may be cheap for a reason the market has worked out and you have not. A workable compromise is to require only that a stock is not in a sustained downtrend, rather than that it leads the market.

5. Liquidity: can I actually trade it?

Average daily traded value, free float, and market capitalisation. This is the filter people forget and then regret. Without a liquidity floor a screen returns microcaps where a modest order moves the price several percent, the spread consumes your first year of expected return, and selling in a bad week is difficult. Set a minimum traded value comfortably larger than the position you intend to take, and keep it in every screen.

How to build a screen for a specific goal

Screens work when they start from an intention, not from whichever fields the tool shows first.

1. Write the goal in one sentence firstFinish this sentence before touching a filter: I am looking for companies that ... . For example, profitable mid-caps growing steadily without leverage at a price that is not stretched. Without the sentence, the screen drifts.
2. Set the universe before the criteriaStart with market capitalisation and liquidity, not valuation. Decide which size band you are shopping in and set your minimum traded value. This removes most of the noise and all of the untradeable names.
3. Add one filter per family, then tightenOne valuation filter, one growth filter, one or two quality filters, optionally one momentum filter, thresholds generous at first. Aim for 20 to 40 names. If 300 match, tighten the quality filters — they are the most informative. If four match, you have over-filtered.
4. Read the list before trusting itSkim every name. If the output is dominated by one sector, one group of holding companies, or businesses with a recent one-off gain, your filters found an artefact. Fix the screen, not the conclusion.
5. Do the real work on five namesTake the five most interesting and read the annual report, segment detail and related-party notes. No screener can do this part, and it is where the decision is made.

Three goals, three different screens

The same tool produces very different screens depending on the intention:

• Steady compounders. Large or mid cap, good liquidity, strong return on capital, low debt, positive five-year revenue growth, valuation as a ceiling. Expect a short list of familiar names.
• Recovery candidates. Depressed valuation plus an improving trend — cash flow turning positive, debt falling, margins off the bottom. Keep quality filters strict; this is where value traps live.
• Momentum for a shorter horizon. A much higher liquidity floor, price above its long-term average, strength versus the index, and a hard exit rule set before you enter.

The mistakes that make screens useless

Over-filtering

The most common failure by a wide margin. Every filter removes companies, and the survivors of a fifteen-condition screen are not the best in the market — only the ones that cleared an arbitrary set of thresholds, often barely. Over-filtered screens also swing wildly when you nudge a parameter, the clearest sign you are fitting noise. If moving a threshold from 15 to 16 empties the list, it was never robust enough to act on.

Screening without ever checking whether the screen works

Most people build a screen, act on it, and never look back. If your tool can test how the filters would have performed historically, use it — sceptically, because historical tests flatter you through survivorship bias, look-ahead bias and quiet tuning. If you cannot test, save today's list with the date and review it in twelve months. Two or three such reviews teach you more than any theorising.

Ignoring liquidity

Worth repeating, because the cost is invisible until you try to leave: illiquid names often look wonderful in a screen precisely because nobody is trading them.

Screening on a single metric

A single-metric screen is a reliable machine for finding one specific kind of bad company. Lowest P/E finds businesses in decline. Highest growth finds companies priced for perfection. Highest dividend yield finds falling prices and payouts about to be cut. Every metric has a characteristic failure mode, and combining families works because those modes rarely overlap.

Screening is not portfolio analysis

These two get confused constantly, and they point in opposite directions. Screening looks outward and asks what should I consider buying. Portfolio analysis looks inward: how much of my outcome depends on my top three positions, how exposed am I to one sector, how far do I swing relative to the market — beta — and how bad has a bad stretch been, which is maximum drawdown. A perfect screen tells you nothing about any of that. It also returns companies that share risks as well as characteristics, so buying a whole list is a fast route to unintended concentration — see diversification for Indian investors.

The two belong in sequence, not in competition. Screen to find candidates; analyse to decide whether adding one improves or worsens the portfolio you have. The common mistake is running the first loop enthusiastically and the second never — buying good ideas one at a time until the portfolio is four versions of the same bet. A screen narrows thousands of companies to a handful worth reading about; it cannot judge management, accounting or competitive position, and it never replaces judgement. Our guides to analysing a stock portfolio and understanding portfolio risk cover the other half.