How to Set a Price Alert That Isn't Noise

The first price alert almost everybody sets is Bitcoin at $100,000. It fires, they look at the phone, they feel briefly informed, and nothing happens. Three weeks later there are eleven more like it and the notifications are switched off.

That failure is not a discipline problem. It is a design problem, and it starts with the number you picked.

An alert is a message from you now to you later. The version of you receiving it will be distracted, on a train, halfway through something else. If the message does not tell that person to do a specific thing, they will dismiss it, and after enough dismissals they will stop reading any of them. So the whole craft of setting a good alert is compressed into one question: at this price, what would I actually do?

Where should the first alert actually go?

At the price where a decision changes. Not at the price you find interesting.

Work backwards from the action. If you have been meaning to add to a position but not at current prices, there is some level at which you would open the exchange and buy. That number is your alert. If you hold something that has run hard and you have quietly promised yourself you would take some off, there is a level at which you would do it. That is your alert.

If you cannot name the action, you do not have an alert. You have curiosity, and curiosity is what the prices widget on your home screen is for. Glanceable information belongs on a widget. Interrupting information belongs in a notification. Confusing the two is the single most common mistake in alert setup.

There is a useful test here. Before you save the alert, write the sentence you want to receive: "SOL just crossed $180, which is where I said I'd trim a quarter." If that sentence sounds absurd when you write it, the level is wrong.

Why do round numbers make bad alerts?

Because everybody else set theirs there too, and price behaves badly around crowded levels.

Round numbers concentrate resting orders. That has two consequences. Price tends to get pulled toward them, so it touches, wicks, and reverses, firing your alert on a move that does not persist. And when a round number does break, it often breaks with force and speed, so by the time you have read the notification and opened an app, the level you were alerted about is already well behind you.

The practical problem is that a round-number alert fires too often and informs too little. Bitcoin will cross $100,000 in both directions dozens of times over a range-bound month. Each crossing is a notification. None of them is news.

Better levels come from structure. The high of a previous range. The low that held three times before this. The price you paid, which matters because it is the level at which your own psychology changes. The price at which a position becomes 20 percent of your portfolio, which is a real risk event even though the chart does not mark it. Reading a candlestick chart on a phone covers how to find these levels without a desktop setup, and support and resistance alerts goes deeper on placing them.

If a structural level happens to sit near a round number, offset it slightly. Set the alert a percent past the round figure rather than on it, so you are alerted to the break rather than to the test.

Above or below — which direction should I choose?

Both directions exist because they answer different questions, and most people set the wrong one first.

An above alert is for something you do not currently own enough of, or for a thesis confirming. It tells you a level has been reclaimed, a range has broken upward, or an asset you were waiting on has stopped being cheap. Above alerts are, honestly, the ones that most often arrive too late to act on. Momentum moves fast.

A below alert is for something you already hold, or want to hold at a better price. It tells you a support level has failed, or the entry you were waiting for has arrived. Below alerts tend to be more actionable, because falling markets give you more time to think than rising ones do, and because the decision behind them is usually already made.

The pairing that works well for a position you own: one below alert at the level where your reason for holding stops being true, and one above alert at the level where you would take something off. Two alerts, two decisions, nothing else. If your reason for holding an asset is not a price level at all, and for long-term holdings it often is not, then skip the below alert entirely and use a percentage alert instead. Price alerts vs. percentage alerts covers when each shape fits.

Why do most first alerts get muted within a week?

Three reasons, and they compound.

The first is volume. Ten alerts across four coins in a volatile week produces a notification every few hours, and human attention treats a frequent notification as background noise within about two days. This is not weakness. It is how habituation works, and no amount of resolve overrides it.

The second is the wick problem. An alert set on the exact touch of a level will fire on a one-minute spike that reverses immediately. You look, see nothing unusual, and learn that the alert was not worth looking at. Learn that four times and you have trained yourself to ignore the app.

The third is that most alerts have no attached decision, which is where this article started. An alert that produces no action is indistinguishable from spam, even when the price information is perfectly accurate.

The fix for all three is subtraction. Fewer alerts, on levels further from current price, each attached to something you would do. A holder with six well-placed alerts gets meaningfully more from them than a holder with sixty. Alert fatigue covers the recovery process if you are already past this point and have muted everything.

How do I set one properly?

Here is the whole sequence, and it takes under a minute once you know the level you want.

1. Decide the action first. Write it in one sentence. Buy, trim, exit, or re-evaluate. If you cannot write it, stop here.
2. Find the level that triggers that action. Use structure from the chart, your own cost basis, or a portfolio-weight threshold. Avoid the round number sitting next to it.
3. Pick the direction that matches the decision. Above for confirmation and for entries you missed. Below for risk and for entries you are waiting on.
4. Offset from the exact level by half a percent or so, so a wick does not fire it. You want to hear about the break, not the touch.
5. Save it, then leave it alone. Moving an alert closer to current price because nothing has happened yet is how noise gets created.

Set two of these, not twenty. Give it a month.

The alert setup that survives is the one you never think about until it goes off, and when it does, you already know what you are going to do. That is the whole standard. If you want to keep going after price levels, the next layer up is being told when volume, market cap or an abnormal move deviates from normal, which is where an alert stops describing a number and starts describing a change in behaviour.