A screener does one job: it turns several hundred listed companies into a shortlist small enough to research properly. Used well, it saves weeks. Used badly, it manufactures false confidence — usually by sorting on one ratio and buying whatever floats to the top.
A sane workflow
1. Start with the question, not the tool. “I want income” and “I want undervalued earnings” are different searches. Income: filter dividend yield above, say, 7% (our screener's High Dividend preset does this). Value: sort by P/E ascending (the Low P/E preset) — then interrogate why each name is cheap.
2. Layer a second filter. A high yield with a payout ratio above 100% is a warning, not an opportunity. A low P/E in a dying sector is a value trap. Combining criteria — yield plus payout below ~80%, or low P/E plus membership of a sector with growing earnings (see our sectors page) — removes most junk automatically.
3. Use sector filters to compare like with like. Banks naturally trade on low P/Es and high yields; consumer names on high P/Es. Comparing across sectors without adjusting for that is how beginners conclude banks are always “cheap” and food companies always “expensive”. Filter to one sector first, then rank inside it.
4. Respect the data. Our own snapshot contained a row implying a dividend yield above 1,300% — a corrupt source figure we removed rather than publish. Any screener, anywhere, will contain some bad rows, stale prices and mismatched periods. If a number looks too good to be true, verify it in the company's own filing before it influences a decision. A dash in our tables means “not in the source” — treat blanks as unknown, never as zero.
5. The screener ends the search; it doesn't end the work. A shortlist of ten becomes two or three only after reading results announcements, checking the dividend record across several years, and understanding what the company actually does. Screening is triage, not diagnosis.
Three mistakes to avoid
Over-filtering: stacking ten conditions until one stock survives — you have curve-fitted, not discovered. Chasing the top of a sort: the #1 yield or #1 return is frequently a data error or a one-off event (our volume table is topped by a Rs 0.10 rights line). Ignoring liquidity: a perfect ratio on a stock that trades a few thousand shares a day is a position you may not be able to exit. Check volume — it is a column in our screener for exactly this reason. Try the workflow now in the PSX screener.