Guide · Updated 5 August 2026

How to Research a Stock in 30 Minutes (Repeatable Checklist)

Most retail research is a random walk through headlines that ends wherever the mood started. This is the fixed sequence we use — the same eight questions, in the same order, every time — so the answer depends on the company rather than on the day you looked.

Why a checklist beats curiosity

Unstructured research has a predictable failure mode: you find the bull case first, it feels good, and everything after that becomes evidence gathering for a decision you already made. A fixed sequence fixes the order of questions so the uncomfortable ones cannot be skipped, and it makes two different tickers genuinely comparable because you asked both the same things.

The eight steps below take about half an hour once you are used to them. They are aimed at holding periods of days to a few months. Nothing here requires paid data.

Step 1 — Say what the company sells, in one sentence

If you cannot describe the revenue engine in a plain sentence — "they design chips and licence the designs to phone makers", "they run regional utilities and sell power under regulated tariffs" — stop. You are about to trade a ticker symbol and a story, and when the position moves against you there will be nothing underneath it to hold on to.

Then add the second half: who pays, and how often. Recurring subscription revenue behaves completely differently in a downturn from one-off hardware sales or commodity-linked pricing. This single distinction explains most of the variation in how sectors behave when the macro tape turns.

Step 2 — Check financial health, not financial beauty

Five numbers, from the most recent quarterly release:

• Revenue growth, year over year. Compare against the same quarter last year, never the previous quarter, so seasonality does not fool you.
• Gross margin trend. Direction matters more than level. Margin compression in a growth story is the earliest reliable warning sign.
• Operating cash flow. A company can report profit and still burn cash. Cash flow is harder to dress up than earnings.
• Net debt versus EBITDA. Roughly above 3x, the balance sheet starts driving the equity story, particularly when rates are high.
• Share count. Rising share count quietly dilutes you. Common in unprofitable small caps that fund operations by issuing stock.

Step 3 — Put valuation in context

A P/E of 40 is meaningless in isolation. Compare it to three things: the company's own five-year range, its closest direct peers, and its growth rate. A business compounding revenue at 30% on expanding margins is not obviously expensive at 40x; a mature industrial growing 3% is not obviously cheap at 18x.

For unprofitable companies, use enterprise value to sales and ask what margin the price is implying at maturity. If the answer requires a margin no one in the industry has ever achieved, the multiple is doing the work, not the business.

Step 4 — Identify the catalyst and put a date on it

A cheap stock with no catalyst can stay cheap for years. Write down the specific event that should force the market to re-price, and when it happens: an earnings date, a product launch, a regulatory decision, a contract award, an index inclusion, a lock-up expiry.

If the only catalyst you can name is "sentiment improves", you do not have a trade with a timeframe — you have a hope with an open-ended holding period. That distinction determines whether a stop-loss even makes sense.

Step 5 — Read the chart for location, not prediction

The chart's job in this process is to answer two questions: is this an area where buyers have historically shown up, and where would the trend read be proven wrong? Concretely:

• Is price above or below the 50- and 200-day moving averages, and are they rising?
• Where are the last obvious swing high and swing low?
• Do RSI and MACD agree with the trend, or is momentum diverging? Our guide on reading RSI and MACD together covers the combinations worth acting on.
• Is relative volume elevated, meaning real participation rather than drift?

Buying into the middle of a range is the most common location error. There is no nearby level to lean the stop against, so every stop placement is arbitrary.

Step 6 — Check liquidity and the mechanics of getting out

Average daily dollar volume is the number that decides whether your exit plan is real. Under roughly $5 million a day, a bad morning can move the price several percent on ordinary retail flow, and stop-losses fill far from where you set them.

Also check the bid-ask spread at the times you actually trade, the size of typical opening gaps, and whether options exist — a liquid options chain usually implies a healthier underlying market. For small caps, add short interest and float: a small float with high short interest can move violently in both directions for reasons unrelated to your thesis.

Step 7 — Write the bear case first

Before you write down why you want to own it, write the three most credible reasons this goes badly: a customer concentration risk, a competitor with better economics, a regulatory exposure, a balance sheet that needs refinancing, a valuation that already discounts perfection.

Then state the falsifier explicitly — the price level or the data point that would make you close the position. "I exit below the July swing low" or "I exit if gross margin declines a second consecutive quarter" are usable. "I'll see how it goes" is not.

Step 8 — Write the thesis in five lines

Compress everything into a note you can re-read in thirty seconds:

Ticker:
What they do:
Why now (catalyst + date):
Entry zone / stop / target:
What proves me wrong:

The discipline is in the compression. If you cannot fit the reasoning into five lines, the reasoning is probably not clear enough to act on. When the position later moves against you, this note is the only thing standing between a planned exit and an improvised one.

Where a scanner fits in

Software is very good at steps 2, 5 and 6 — pulling financial figures, computing technical structure, and measuring liquidity across hundreds of tickers in seconds. It is weaker at step 1 and step 7, where judgement about a business model and its risks matters. Use the machine to narrow a universe down to a handful of candidates, then spend your half hour on the ones that survive.

That is exactly how SIGNAL9 is built: an automated pass over the universe produces a ranked shortlist with the technical and liquidity work already done, and the written thesis tells you what the model thinks the catalyst is so you can disagree with it. How that pipeline works, and its limits, is documented in our editorial policy and methodology.

The one-page version

1. What do they sell, and who pays?
2. Growth, margin trend, cash flow, debt, share count.
3. Valuation versus own history, peers and growth.
4. Catalyst — and its date.
5. Chart location and trend agreement.
6. Liquidity and spread.
7. Bear case and falsifier.
8. Five-line written thesis.

Run it the same way every time. The value is not in any single step — it is in the fact that you cannot skip the inconvenient one on the day you happen to be excited.

FAQ

Disclaimer: educational content only, not investment advice. See our disclaimer.