How to Monitor Stocks Without Reading the News Daily

How do you monitor stocks you own without living in the feed? Write down two to five assumptions your view of the business rests on, give each of them one observable signal and a date on which you will check it, and return to that page on the rhythm of quarterly reports. The rest of the stream is weather. Monitoring that starts with a headline instead of an assumption is not monitoring, only reading news that happens to contain your ticker.

A tracker is not a monitor#

The standard answer to this question is an answer about tools. Connect the account to an app that pulls quotes. Build a sheet with cost, weight and return. Set alerts on the ticker, check the numbers once a quarter, rebalance once a year. It is sensible advice and half of it genuinely works. Without a weights table you will not notice that one company has grown into a third of the portfolio. Without an earnings calendar you miss the one day when there is anything to read at all.

Which is portfolio accounting, not company monitoring.

Accounting answers how much you hold and how much you have made on it. Monitoring answers something else: whether the reason you hold it is still true. The two questions take different inputs and run at different speeds. A valuation changes every second of the session. A business changes a few times a year, and almost always when the company itself publishes something. A portfolio in this sense is the list of companies you keep watch over rather than a balance at a broker, which is exactly why a portfolio is not your brokerage account.

The weakest part of the standard kit is the ticker alert. It does not react to a fact, it reacts to a string. A post listing your company among "five stocks for autumn" arrives at the same priority as a filing about the loss of the largest customer, because both contain the same three letters. A filter that cannot tell those apart manufactures vigilance without content, and within two months it teaches you to ignore your own alerts. Worse than no filter, because it feels like coverage.

What your view of the company rests on#

Start with the sentences holding the structure up. An investment thesis is not an opinion about the price. It is a handful of statements about the business that must be true for the rest of your reasoning to stand. There are few of them because only a few things carry weight. The test for each is the same: if it turned out to be false, would you have to rebuild your picture of this company? If not, it is not an assumption. It is decoration.

Say a maker of cables for data-centre racks. Its story rests on three sentences. The data-centre segment is growing faster than the rest of the company and has lifted its share of revenue from 20 to 40 percent in two years. The new standard promised for the second half of the year will actually reach volume production. Gross margin, which has slid from 25 to 21 percent over four quarters, will stop falling once two product generations stop overlapping. The numbers are round because they are invented for the example, and the shape of the list is real.

Which means the whole story of this company fits into three sentences, each of which a document can settle. Thirty headlines in the same month will settle none of them.

One warning about the first list you write. An assumption is not a risk and not a fact. A fact ("the segment is 40 percent of revenue") has already happened and needs recording, not watching. A risk ("a fire at the plant") is possible but does not sit underneath your reasoning, so it belongs in a different column. An assumption is a statement about the future on which your answer to "why do I hold this at all" depends, and only those go on the page.

Two assumptions are usually too few to describe a business, and six is a sign you are mixing what carries weight with what is interesting. When the list grows, ask of each sentence what changes if it proves false. The sharper version of this exercise, run from the demolition side, is laid out in how to break your own investment thesis.

Every assumption gets a signal and a date#

An assumption without a signal is a wish. A signal is something you can open and read: a line in the segment table, a dated statement from management, the gross-margin line. It is not "improving sentiment in the industry", because that cannot be settled, only leaned on.

The second column has one trap in it. A signal has to live in the company's own document, not in commentary about the company. A segment's share of revenue sits in a table and either rose or did not. "Growing interest in data centres" sits in somebody else's article and always rises, because such articles get written when it does. If your signal can be confirmed by three opinion pieces and no line in a filing, it is not a signal. It is an echo.

The last column matters more than it looks. Without a date you check the company when you are uneasy, and unease correlates with the price rather than with the business. I think adding a deadline to each assumption changes more than the assumption itself, because it converts vigilance that arrives in waves into a review that sits in a calendar. You check the reserve chute on the ground, in daylight, when nobody is panicking.

What you deliberately ignore#

The second list is shorter and less pleasant to write. It is the inventory of things you decide in advance not to react to. A price move with no event underneath it. A broker changing a rating, a piece about "sentiment" in the sector, a competitor's ad campaign, a headline repeating a number that has sat in a filing for two weeks.

There is one rule. If an event settles none of your sentences, it is not information about your company, only weather around it.

That list is necessary because the stream is built for attention, not for decisions. The stock moved two percent, so somebody wrote nine hundred words about why. The honest answer was usually that it moved two percent. A longer version of that argument sits in the piece on noise versus signal in investing.

The same arithmetic decides how many companies you can actually watch. Three assumptions times four quarters is twelve checks a year per company, each a quarter of an hour. At five companies that fits into one afternoon a quarter. At twenty it fits nowhere, and you end up watching whichever one is falling loudest. Attention is the binding constraint here, not access to data. You have more data than you can read and the same number of afternoons you always had.

The point is not to stop reading. The point is to stop counting reading as monitoring.

The filing calendar sets the pace#

You do not need to invent a rhythm, because somebody has already set one for you, and it was not an editor. A large accelerated filer in the US has 60 days from the end of its fiscal year for the 10-K and 40 days from the end of the quarter for the 10-Q (deadlines from the SEC rule accelerating periodic reports). An event the company judges material has four business days to reach a Form 8-K (the form's general instructions).

Which means that in a normal year you get four dates on which verifiable facts genuinely accumulate, plus a handful of irregular filings in between. Four documents a year against a stream that produces something every hour: that is the whole disproportion this piece is about.

Four times a year the company has to tell you something, and the rest of the time somebody else is talking about it.

Deadlines differ in other jurisdictions, but the principle does not. The publication calendar runs slower than the newsroom calendar, and it decides when there is anything to check. Your list inherits four ready-made dates per company. The only thing you add is the decision about what you look at on those days.

Put a date on the first note#

When you write the assumptions down, put the day they were written at the top. It looks like a formality and it is not. The date turns a note into a reference point, because from then on every later reading answers a smaller question: what has changed since last time?

That question is cheaper than "how does this company work". On the second reading of an annual report you do not read everything with the same attention. You go to the business description, the risk factors, the segment breakdown, the demand commentary and management's language, and you look for the places that have moved relative to your page. The rest stays where it was.

Most of the value sits in the risk section read year over year. The company writes it under legal compulsion, so it rarely adds anything there for decoration, and every new sentence means the lawyers judged that thing real enough to name. A paragraph that disappears can say just as much. Without a dated page you have nothing to compare against and you read the section as if for the first time, which loses the only information it really carries: the difference.

An investor's memory is very kind to its owner. After the fact almost everyone knew the risk was there, and almost nobody has evidence. A dated page is that evidence, and incidentally the only way to check a year later whether you watched what mattered or what was loudest. I do not know how many of your current assumptions will survive twelve months. I do know that without a date there is no way to settle it.

Three sentences after every quarter#

Results are the only moment in the quarter when your list can genuinely change. So end each of those readings the same way, with three sentences written under a date:

1. Confirmation. Which assumption got evidence, and from exactly where.
2. Warning. Which one moved against you, even slightly.
3. Question. What you still do not know, and where you will look next time.

For our cable maker it might read roughly like this. Confirmation: the data-centre segment's share rose for another consecutive quarter and it is in the segment table. Warning: gross margin fell another half point and management explained it by overlapping product generations for the second time running. Question: when does a concrete volume-production date for the new standard appear, because this quarter again it did not. Three lines, one date, no adjectives.

Three sentences take five minutes and do something no amount of reading does: they leave a trace. After four quarters you have a sequence that shows whether an assumption is strengthening or merely has not been refuted yet. Those are two different things, though up close they look identical.

Do not do it on the day of publication. The numbers land in the evening, the call runs an hour, and the first coverage is written before anybody reaches the Q&A. A day of delay costs you nothing and buys the full transcript and the analyst questions, which is the same reasoning behind why Taufolio waits a day after earnings.

When the price falls and the assumptions hold#

What breaks in this method? It has two weak points, and they are better named before any conclusion.

The first: a list of assumptions is a hypothesis about what matters, and it can simply be wrong. You picked margin, and the company was upended by a customer concentration you never thought about. The answer to that is not a longer list but an annual review of it. Once a year you check which of your sentences actually explained the changes and which one sat empty through four quarters.

The second is worse because it is comfortable. A list that keeps holding can turn into an alibi. The price falls by a fifth, you look at three undisturbed sentences and draw from them a calm the data has not yet earned. Price is information about what other people know, so a sharp fall is a question rather than an answer: check whether one of your assumptions has just been settled, and only then go back to being calm. How to walk through such a move step by step is set out in what to do when a stock drops 20 percent.

Who keeps the list#

The list itself is cheap.

Keeping it alive for twelve months is expensive, because it means remembering four dates per company, where the page is, and that the risk section has to be compared with last year's version rather than read afresh. That is usually where the method dies: not for want of an idea, but on the third busy month in a row.

In Taufolio that page is called the Investment thesis and it comes out of a Full report automatically, broken into a few assumptions with a measurable condition on each. The rhythm is held by Monitoring. That covers the thesis check after results, a Pre-earnings brief before them, a Monthly summary with The gist of the month at the top, and The gist of the news beside the stream that flows past anyway. Breakthrough news fires only when something happens that could change the company fundamentally. None of it costs Credits and all of it works on every plan, the free one included. The Free plan covers one company under Monitoring. Alerts, conditions written in your own words, sit separately, and the Free plan does not have them. How the pieces fit together is easiest to see on the product page.

If you are going to do one thing today, do not start with the list. Start with the dates. Put the four reporting dates of every company you hold into a calendar, and at the first of them write the sentences you intend to watch. A list with no date sits in a drawer. A date with no list will force you to write one.