Building a calmer news diet

Most news consumption is not informational

It is worth being honest about the mechanism. A large share of time spent checking news is not an attempt to learn something specific; it is a response to unease, and it produces relief for a few seconds followed by more unease. The feed refreshes. Understanding does not accumulate.

The evidence for this is simple: after a long session, most people cannot state what they learned. Compare that to reading one substantial article, which reliably leaves something behind. The problem is not the amount of information available. It is the absence of any rule governing when to look and what for.

Time-box, and give the session a question

Two constraints do most of the work. First, decide when you look — for instance twice a day, fifteen minutes each, at times that are not the first and last thing you do while awake. Second, decide what you are looking for before you start, written as an actual question: what changed overnight that affects my work, my city, or the few companies I follow?

A question is what makes a session end. Open scrolling has no completion condition, which is why it expands to fill whatever time is available. With a question, you either have your answer or you establish nothing happened, and both are finished states.

The corollary is to turn off breaking-news notifications for everything you cannot act on. A notification is a claim on your attention made by someone optimizing for engagement, and almost nothing genuinely requires you to know within the hour.

Choose beats, not brands

People tend to organize their reading around publications, then feel obliged to consume everything those publications produce. Organizing around beats works better, because quality is beat-specific: an outlet may be excellent on courts and unreliable on semiconductors, and the reverse is true of another.

Write down the four or five beats you actually need — your industry, local government, a couple of companies, one international desk — and identify who does each well. Everything outside that list becomes explicitly optional, which is the point. Optional is not the same as ignored; it means it reaches you later, through a weekly digest or someone who covers it properly, rather than continuously.

Separate the alert from the understanding

These are different activities and conflating them is what makes news exhausting. Scanning headlines tells you what exists. Reading tells you what happened. Trying to do both simultaneously produces a state of knowing a great many story titles and very few stories.

Split them deliberately. Scan quickly, note two or three items worth actual attention, and read those properly — ideally later, when the initial coverage has been corrected and the primary documents are available. Almost every story is better twelve hours old, and the small number that are not are usually the ones you would hear about regardless.

Prefer documents to coverage where you can

For anything that matters to you specifically, the primary material is usually available and usually shorter than the coverage: the filing, the ruling, the bill text, the press release, the research abstract. It is drier, harder to share, and much harder to distort with a spicy framing.

A useful personal rule: no consequential decision from a headline alone. If a story would change how you act — at work, with money, about your health — the cost of ten minutes with the source document is trivial against the cost of acting on a distorted retelling.

Notice when you are being farmed for reaction

Some content is engineered to produce a feeling in the first two seconds, because the feeling is what causes sharing. Outrage, fear, and urgency are the three most common. This is not a claim about any particular outlet's intentions; it is a description of what the incentive structure of engagement-based distribution selects for.

The practical response is to treat a strong immediate emotional reaction as a signal to slow down rather than to act. Ask what the sender gains when you feel this way, and whether the underlying event is as extreme as the framing. Frequently the event is real and ordinary, and the framing did the work.

The weekly audit

Once a week, spend three minutes reviewing your inputs. This is the step that makes the rest durable, because it converts vague dissatisfaction into a specific change.

• Which two or three things did I read this week that I still remember and could explain to someone?
• Which sources produced stress without ever changing a decision?
• Did I confuse being early with being informed?
• How many times did I check outside my intended windows, and what triggered it?
• What did I read a summary of that deserved the primary document instead?

Then remove one input. Almost everyone's list has something on it that survives purely by habit, and removing it costs nothing measurable.

What a good week looks like

Not abstinence. A reasonable target is a couple of short scanning sessions a day, a handful of articles read properly rather than skimmed, one or two primary documents on anything that affects you directly, and a weekly longer piece on something outside your immediate concerns.

That is a modest amount of reading, and it produces substantially more understanding than several hours of scrolling, because the constraint forces selection. The goal was never to consume less news. It was to end the week knowing something you did not know at the start of it.