How to Use LyraAlpha Watchlist Alerts Without Getting Overwhelmed
The most common mistake with watchlist alerts is setting too many of them.
When everything is an alert, nothing is an alert. You start dismissing alerts because you cannot keep up, and eventually you either turn them all off or you stop noticing them entirely. The system that was supposed to keep you informed becomes noise.
Here is how to configure alerts that stay useful.
The Alert Quality Principle
An alert is useful if it meets two criteria:
- It fires when there is something you would actually change a position or thesis based on
- It does not fire when there is nothing actionable to do
An alert that fires 20 times a day and requires you to dismiss 19 of them is not keeping you informed — it is consuming your attention for no benefit.
The goal is not to be alerted to everything that happens. It is to be alerted to the things that matter.
How to Configure Alerts by Signal Type
Tier 1: Regime-Level Alerts
Set these and never turn them off. They are the most important signals and they fire rarely enough that they never become noise.
Market regime shift alert: Fire when LyraAlpha's overall market regime classification changes. This is the highest-level signal in the system. When the regime shifts, your entire portfolio context changes and you need to know.
Correlation regime shift alert: Fire when the overall cross-asset correlation regime changes from normal to elevated or crisis levels. This tells you whether the market is in differentiated or indiscriminate mode.
Tier 2: Position-Level Alerts
Configure these for your top 5-10 positions. These are the assets where you would act if a significant signal fired.
On-chain flow alerts: Configure for significant exchange inflow or outflow events — typically 2-3x the 30-day average. This tells you when something material has changed in the supply dynamics of an asset you hold.
Regime alignment change: Fire when an asset's regime alignment changes — when it stops being aligned with the bull regime and starts behaving like it is in a different regime. This tells you when the market's perception of an asset has shifted.
Tier 3: Opportunity Alerts
These are optional and should be configured only when you are actively researching or considering a position. When you are not actively interested in an asset, turn these off.
Signal discovery alerts: When you are researching an asset, set alerts for the signal types most relevant to your thesis. When the research is done or the thesis is resolved, turn the alerts off.
The Maximum Alert Budget
A useful alert budget for an active investor managing 10-20 positions:
- 2 regime-level alerts (always on)
- 10-15 position-level alerts (Tier 2)
- 5-10 opportunity alerts (temporary, rotating)
Total: 17-27 active alerts maximum.
If you have more than 30 active alerts, you have too many. Cut the Tier 3 opportunities first, then reduce Tier 2 to only your top 5 positions.
What to Do When an Alert Fires
When an alert fires, the response protocol is:
- Read the signal context: Why did the alert fire? What regime is the market in? What does the signal mean in that context?
- Assess materiality: Does this signal, in this regime, warrant a position change? Would you change anything if you had this information?
- Act or dismiss: If the answer to #2 is yes, act. If the answer is no, dismiss with a note if the signal was borderline — this data improves future filtering.
- Do not escalate: If the signal is not actionable, do not escalate it to a position change out of anxiety. Wait for the next data point. Alerts are signals, not commands.
Alert Anti-Patterns
Setting alerts at every threshold: Alerting on every 5% price move, every small exchange flow, every minor funding rate shift. This is noise, not signal.
Never dismissing or reviewing: Setting alerts and never coming back to evaluate whether they are producing useful signals. Alerts should be reviewed monthly and adjusted based on signal quality.
Only alert on downside: Setting price-drop alerts but not price-breakout alerts. You need to know when positions are working too, not just when they are failing.
Setting and forgetting: The same threshold forever. Market conditions change and alert thresholds should be recalibrated quarterly.
How LyraAlpha's Alert Intelligence Works
LyraAlpha's alert system is regime-aware by default. This means alert thresholds are automatically adjusted based on current market conditions.
A 10% price move in a high-volatility regime is less significant than the same move in a low-volatility regime. LyraAlpha accounts for this automatically — the alert threshold is relative to the regime, not an absolute number.
This is why the default alert configuration in LyraAlpha is already calibrated for signal over noise. The system is designed to surface high-confidence signals and suppress noise in high-volatility regimes.
The user's job is to configure which signal types they care about — not to tune absolute thresholds manually. Let the regime-aware system handle the threshold calibration.
FAQ
Q: How do I know if my alerts are too sensitive?
A: If you are dismissing more than 70% of alerts without acting on them, your thresholds are too sensitive. The signal-to-action ratio should be roughly 50-50 — about half of fired alerts should lead to a review and some subset should lead to a position change. If nothing is ever actionable, you are in alert noise.
Q: Should I set alerts on assets I am watching but do not own?
A: Yes, but in the opportunity tier, not the position tier. Alerts on non-owned assets should be temporary — set them when you are actively researching a potential addition, keep them for 2-4 weeks, then either act on the thesis or close the alert track and move on. Do not maintain long-term alerts on a watchlist of 30 assets you do not own — that is monitoring, not intelligence.
Q: Does LyraAlpha have smart alert filtering?
A: Yes. LyraAlpha automatically adjusts alert thresholds based on the current regime. In a high-volatility regime, absolute thresholds are less sensitive because the system recognizes that elevated volatility produces more signals that are noise. In a stable regime, the thresholds are tighter because signals are more likely to be meaningful. This regime-aware filtering is built in and automatic.