When revenue dips, you hear first
Custom alerts in Slack and email
A morning KPI report, a ping when ROAS drops below target, a heads-up when a metric moves out of its normal range. The numbers come to you before you open a dashboard. Build alerts in four steps, or just ask Claude or ChatGPT.
Three kinds of alerts, built in four steps.
Pick what to watch, when to alert and who gets it, check the preview, and save.
- 1Name & type
- 2Target metric
- 3Condition
- 4Recipients
Pick the alert that fits the moment.
Send a report on a schedule, or only when a condition is met.
Scheduled report
“Every day at 9 AM, send yesterday's revenue, orders and AOV vs. last week”
- Daily or chosen weekdays at a set time
- Up to 5 saved charts in one alert
- Choose whether to send when there is no data
Metric condition
“Tell me when ad ROAS falls below 180%”
- Above, below, equal, up or down vs. the previous period
- Up to 10 conditions, all or any
- Each condition reads as a plain sentence
Anomaly detection
“Tell me when payments move outside their normal range”
- Learns trend and weekday patterns to set an expected range
- Low, normal or high sensitivity
- Same engine as anomaly detection on dashboard charts
In the app, it takes four steps.
Each step starts with sensible defaults, so you only change what you need.
Pick data directly, load a chart you already saved, or duplicate an existing alert.
Send on a schedule, or only when a metric condition or anomaly is detected.
Choose Slack channels or email recipients and the delivery days and time. You can reuse the delivery settings of your last alert.
See a preview built from live data and send yourself a test before saving.
Or just ask Claude or ChatGPT.
Connect Laplace over MCP and create, edit, pause or resume alerts from the AI chat you already use. The AI validates the setup and shows a preview before saving, so all you do is confirm.
MCP is available from the Growth plan and during the free trial.
Learn about MCPI grouped revenue by channel and ROAS for yesterday. The setup is valid and ready to save.
Preview with live data: yesterday's revenue $11,200 · Amazon 52% · ROAS 312%.
Created “Daily revenue report”. The first one goes out tomorrow at 9 AM.
| Say this | What the AI does |
|---|---|
| “Every Monday at 10, email me last week's top 10 products by contribution profit” Validate → preview → create scheduled report | Validate → preview → create scheduled report |
| “Tell me in Slack if yesterday's revenue is below $7,500” Check condition → create metric alert | Check condition → create metric alert |
| “Show my alerts and turn off the ones I don't use” List alerts → pause | List alerts → pause |
| “Add my manager's email to the “Weekly revenue” alert” Read current setup → update recipients only | Read current setup → update recipients only |
| “Why didn't yesterday's alert arrive?” Read delivery history → explain the failure | Read delivery history → explain the failure |
Delivered where your team already is.
| Channel | Best for | Note |
|---|---|---|
| Slack | Seeing the same numbers together in a team channel Connect your workspace and pick channels | Connect your workspace and pick channels |
| Reading several tables and charts in detail Admins can block external email addresses | Admins can block external email addresses |
After it's sent, nothing slips by.
Delivery history
Every alert keeps a record of when it went out, to whom, and whether it succeeded or failed, with the reason.
Send now
Send once right away with the saved settings, without waiting for the next scheduled time.
Flags chart changes
If a metric in the chart an alert watches changes or disappears, the alert is flagged and links to the fix.
Waits for late data
If data collection runs late, delivery waits up to 6 hours so an empty report doesn't go out.
Pause and resume
Turn an alert off for a while without deleting it, and turn it back on later.
External email policy
Admins can limit email delivery to project members and approved domains.
Custom alerts, frequently asked questions
Still have questions? Contact us anytime — we will tailor the guidance to how you plan to adopt Laplace.
When you sign up, a default alert that sends key metrics every morning is created for you. Custom alerts let you choose the metrics, conditions, recipients and time yourself, as many as you need.
It learns the trend and repeating patterns such as weekdays over the comparison period, calculates an expected range, and alerts when the actual value falls outside it. You can choose low, normal or high sensitivity. It uses the same engine as anomaly detection on dashboard charts.
A scheduled report can include up to 5 saved charts. Metric and anomaly alerts evaluate a single chart.
Yes. Connect Laplace over MCP and say something like “Send yesterday's revenue to Slack every morning at 9” to create an alert, change recipients or time, or pause it. The setup is validated and previewed before saving.
You choose the days and time. Alerts are not sent between 00:00 and 08:00 KST while data is being collected.
Each alert's delivery history shows successes, failures and the reason. Alerts that stopped because their chart changed are also flagged in the list.