How to Build a Stock Watchlist From a Screen

Your screener returns 140 stocks. Saving all of them as a watchlist feels productive and achieves nothing, because a list that long never gets read. To build a stock watchlist from a screen, start with a written question, filter for liquidity before anything else, add the filters that answer the question, and cut what is left to a list you can read every time you open it. Every name on it gets three notes: why it is there, the price where you would act, and when the company next reports. Then decide how often the screen runs again, and what makes a name leave.

Write the question first

A watchlist answers one question. Which dividend payers have dropped toward a yield worth buying at? Which growth stocks are pulling back to a moving average? Which small caps have rising volume ahead of results? Each question needs different filters and different columns, so a list without one turns into a pile of tickers that nobody can act on.

Write the question in a sentence. Put it at the top of the list.

Set liquidity minimums before any other filter

The first filters keep out stocks you could not trade cleanly: a floor on share price, a floor on average daily volume, and a floor on dollar volume. Setting liquidity minimums first explains the reasoning. A cheap-looking stock with only a few thousand shares changing hands each session can cost more in the spread than the move you were hoping for.

Dollar volume puts a 5 stock and a 500 stock on the same scale. Set the floor according to the size of positions you trade. Bigger positions need a higher floor.

Add the filters that answer the question

Now add the filters that match the question, a few at a time. Run the screen after each one and watch how the count changes. A filter that cuts the list from 140 to 3 is probably too tight. One that changes nothing is doing no work and can go. Using a stock screener covers filter types and how different screeners define the same field.

Save the screen with its exact settings, and write the date next to them, since a screen you cannot rerun with the same filters next month cannot tell you which names are new and which have dropped out.

Cut the results to a list you will read

Screens rarely return a short enough list on their own. Cut the rest by hand. Open each chart. Read the latest earnings headline. Drop anything you would not trade even if it hit your level, whether because of an upcoming event, a business you do not understand, or a chart that tells a different story from the numbers.

What survives should be short. Building a watchlist you will actually use covers keeping it that way.

Give every name a reason, a level and a date

Each row needs three things in writing.

• A reason. One line on why it is on the list.
• A level. The price where you would act.
• An earnings date. The next scheduled report.

The earnings date matters more than it looks. A stock that reaches your level the day before results is a different trade from the same stock a month earlier, because the report can move the price well beyond any level on the chart, and you want to know that before you place the order. Quote pages usually show the next earnings date. Check it against the company’s investor relations page closer to the time.

Choose columns that show when to act

Pick columns that answer the question at a glance:

Column	What it tells you
Last price and your level	How far the stock is from action
Distance to level, in %	The same thing as one comparable number
Relative volume	Whether today’s trading is unusual
RSI	Whether the move is stretched
Next earnings date	Whether an event is close

Sort by distance and the names closest to action rise to the top. Relative volume shows whether interest is building, and the volume bars under the chart show how it built. The RSI column is a check on how stretched the move is. Neither is a signal on its own.

One question, start to finish

Take a hypothetical question: which dividend payers are trading near a yield worth buying at, without paying out more than they earn? The thresholds below are illustrative choices for the example, not recommendations.

Each of the eight gets a row: a reason such as a dividend raised for several years running, a level where the yield would reach the target, and the next earnings date, since a dividend cut is most often announced with results and that is the day the reason on the row could stop being true.

Short-term and long-term lists

The same method builds both, with different settings. A short-term list leans on relative volume, distance to level and the RSI column, and gets rerun every week, sometimes every day. A long-term list leans on valuation and dividend columns, carries fewer names, and can be rerun monthly. Mixing the two on one list puts a stock you might hold for years next to one you might hold for days, and the columns that matter for one are noise for the other.

Keep them apart. Name each list after its question.

Re-run the screen and retire names

This is the step that keeps a watchlist honest, and the one most often skipped. Pick a schedule, weekly for a short-term list and monthly for a longer one, and run the saved screen again on that day. Then compare.

• Names still passing stay.
• New names that pass get the same review as the first batch.
• Names that no longer pass leave the list.
• Names that hit their level move to a separate list for orders.

Write down why each name left. After a few rounds, the notes show whether the screen keeps finding stocks that do what you hoped, which is the only real test of whether the filters were any good in the first place.

When a rerun returns nothing, resist loosening the filters straight away. An empty screen is information. A yield screen can go quiet after a long rally lifts prices and pushes yields down, and a pullback screen can go quiet in a steady uptrend, so an empty result may simply mean the market is not offering what the question asks for right now. Loosen one filter only if the question itself has changed. Otherwise wait for the next run.

What goes wrong

• Saving every result as a watchlist.
• Filtering for value or growth before liquidity.
• Losing the screen settings, so it can never be rerun.
• Forgetting the earnings date.
• Keeping names that no longer pass the screen.