Events
Provide Events data in Alviss AI for time-bound occurrences like Force Majeure strikes outages to model disruptions on operations sales KPIs.
Events data captures time-bound occurrences that may impact business operations, such as Force Majeure events (e.g., natural disasters), strikes, website outages, or other extraordinary circumstances. These events span a defined period and can be used to model disruptions or anomalies in sales, demand, or other KPIs.
Sourced from internal logs, news reports, or operational records, this data allows Alviss AI to incorporate event-based controls in modeling. For example, it can quantify the impact of a system downtime on revenue, helping to isolate external factors in attributions, simulations, predictions, and optimizations. By including events, models can account for non-recurring influences, improving accuracy and providing insights into resilience strategies.
Data Requirements
The Events Data file must include the following columns (headers). All columns are required unless marked as optional:
- Country (string, required): A three-letter country code per ISO 3166 standard (e.g., "SWE" for Sweden), indicating where the event occurred.
- Region (string, required): The region within the country (e.g., "all" for nationwide or a specific ISO 3166-2 code like "SE-AB" for Stockholm County in Sweden).
- Grouping (string, optional): For additional segmentation, such as product categories, demographics, or sales territories (e.g., "all" if not applicable).
- StartDate (date, required): The start date of the event in ISO 8601 format (YYYY-MM-DD).
- EndDate (date, required): The end date of the event in ISO 8601 format (YYYY-MM-DD). This can be the same as StartDate for single-day events.
- Event (string, required): The name or identifier of the event (e.g., "Website down", "Strike"). The same event can span multiple rows if it affects different locations or groupings.
Supported Formats
Data can be uploaded in long (tidy) or wide (pivoted) format. Choose based on your data pipeline:
- Long Format (Recommended for flexibility): Each row represents a single event instance (one event per location, grouping, date range).
| Country | Region | Grouping | StartDate | EndDate | Event |
|---|---|---|---|---|---|
| SWE | all | all | 2018-01-08 | 2018-01-10 | Website down |
| UKR | all | all | 2019-10-21 | 2019-10-24 | Strike |
- Wide Format (Useful for spreadsheets): Columns represent combinations of Country, Region, Event, and Metric (StartDate/EndDate), with rows containing the date values. The first few rows define fixed attributes.
| Country | SWE | UKR | SWE | UKR |
|---|---|---|---|---|
| Region | all | all | all | all |
| Event | Website down | Strike | Website down | Strike |
| Grouping | all | all | all | all |
| Metric | StartDate | StartDate | EndDate | EndDate |
| 2018-01-08 | 2019-10-21 | 2018-01-10 | 2019-10-24 |
Best Practices
- Consistency: Ensure dates align with your project's periodicity (e.g., weekly events if the project is set to weekly granularity). Overlapping or mismatched dates may require aggregation.
- Data Quality: Verify event dates for accuracy and completeness. Use the Activities dashboard to cross-reference with sales or other data for validation.
- Granularity Alignment: All data in a project must match the chosen periodicity (set during project creation). For details, see Projects.
- Event Naming: Use clear, consistent event names to enable easy filtering. Categorize events (e.g., "Operational: Website down") for advanced grouping in models.
For uploading instructions, see Upload Data. If you encounter issues, contact support or refer to the API for programmatic uploads.