Hello, I have a quick question and I was wondering if somebody could help. I am working in the data governance team at veepee and we have recently started using elementary. I am currently in the process of setting up volume anomaly testing on all of the raw tables that we have to ingest data. I want to incorporate the seasonality parameter into my volume anomaly tests at the seasonality: day_of_week level. However, when I do this I am told that I need to have my time bucket at day level. My problem here is that since elementary is only running tests on data where the time bucket needs to be completed - in the case that we have a volume anomaly we will only capture it the day after it occurs - but, we want to capture the anomaly the day it occurs. Any suggestions on how to get around this issue? Below is an example test we are using. Thanks in advance

tests:
      - elementary.volume_anomalies:
          timestamp_column: "meta_processing_time" # timestamp field to group the volum
          tags: ["volume_freshness"]
          days_back: 14 # nb of days of data to retrieve
          anomaly_sensitivity: 3 # 1 lowest - 3 highest 
          anomaly_direction: drop #spike: Increase of volume | drop: decrease of volume | both: spike & drop
          backfill_days: 2
          time_bucket:  
            period: day
            count: 1
          seasonality: day_of_week
          config:
            severity: warn # warn | error

Hey @Luke CLARKE, sorry for the late response 😅 🙏 the configuration you wish for is not supported, as elementary needs the data of the full day to determine the anomaly, you might want to combine multiple tests for full coverage (hourly threshold for immediate issues, and day of week tests for broader issues)