[Deprecated] Continuous Scan
Overview
Continuous Scan is a parameter that enables alerting on data quality issues that occur between subsequent monitor runs. When you turn it on, it automatically includes all anomalies that happened since the last monitor run in the next incident. It prevents gaps in monitored data, for example when the run schedule changes frequently.
How To
Activation

Turn on the setting in the Monitor Editing screen. It requires no further parameters.
Availability
Continuous Scan is available for machine learning (ML)-based Sifflet monitor templates. It is available through the UI only and is not supported in Monitors as Code.
Use Cases
Example #1 — Misalignment Between Data Aggregation and Schedule
Consider a monitor with the following parameters:
- Template: Metrics (Dynamic) >> Normalized Average
- Time Window (Model Training Period): 365 days
- Time-based Data Aggregation: hourly
- Schedule: @daily
These settings result in 1 data point per hour, created every day.
- With Continuous Scan OFF, the monitor raises alerts only on data quality issues concerning the last created data point (in this case: occurring on the hour of the run, potentially missing 23 anomalies).

Monitor run from January 5, 2024, reporting only on 1 anomaly on the last hour of the daily run.
- With Continuous Scan ON, the monitor raises alerts on data quality issues occurring on any data point (in this case: 24 data points, 1 per hour) created during a daily monitor run.

Monitor run from January 5, 2024, reporting anomalies in 9 timeslots across the last 24 hours.
Example #2 — Changing Time Offset
Context: After data pipeline updates, data now arrives with a 1-day delay instead of a 7-day delay. Therefore, you adjust the monitor Time Offset setting as follows:
- Template: Metrics (Dynamic) >> Normalized Average
- Time Window (Model Training Period): 365 days
- Offset: 7 days -> 1 day
- Time-based Data Aggregation: 1 day
- Schedule: @daily
These settings still create 1 data point per day, but it now starts 1 day before the run date (T-1) instead of 7 days before the run date (T-7).

Monitor run from January 5, 2024, with a 7-day offset, showing no anomalies on the analysis day (December 29, 2023).
- With Continuous Scan OFF: during the first monitor run with the new schedule, there is a window of unmonitored data of 6 days (from T-7 to T-1).

Monitor run from January 5, 2024, with a 1-day offset, reporting only 1 anomaly — on the analysis day (January 4, 2024).
- With Continuous Scan ON: during the first monitor run with the new schedule, that run covers the window of 6 days (from T-7 to T-1), as shown in the screenshots below.

Monitor run from January 5, 2024, with a 1-day offset, reporting anomalies in 7 timeslots.
Updated 25 days ago

