Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.datagenie.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
What Filter Presets is

Filter Presets are story-generation controls, not reporting filters. They shape which autonomous stories DataGenie creates in the first place — narrowing the search space, applying qualification rules, and prioritizing the signals that match your team’s analytical focus. Once you’ve identified which segments, KPIs, or dimension combinations matter most, save them as a Preset and every subsequent Top Stories run respects that scope.

​
What you can configure

KPI Filter

Define which KPIs should be considered during story generation — including view filters, min/max thresholds, impact filters, and absolute deviation rules.

Dimension Filter

Control where DataGenie should look for anomalies — select dimension groups or specific values using AND/OR logic.

Sub-population Filter

Ensure statistical relevance by filtering out low-volume segments and preventing misleading anomalies from small sample sizes.

Depth Filter

Control how deeply dimensions are combined during analysis — lower depth for broader trends, higher depth for granular insights.

​
KPI Filter

Define which KPIs should be considered during story generation.
  • View Filter — Ensures selected KPIs are always included in the evaluation scope and increases their visibility in Top Stories.
  • Min / Max Thresholds — Restrict KPI values to specific ranges using actual or predicted values.
  • Impact Filter — Surface only anomalies with specific impact levels (e.g., critical positive or negative).
  • Absolute Deviation — Filter out low-signal changes by enforcing minimum deviation from the predicted baseline.

​
Dimension Filter

Control where DataGenie should look for anomalies.
  • Select dimension groups (e.g., Region, Channel) or specific values.
  • Use AND / OR logic to define how filters combine.
  • Optionally restrict contributors to keep stories focused and actionable.

​
Sub-population Filter

Ensure statistical relevance by filtering out low-volume segments.
  • Focus analysis on high-volume data segments.
  • Prevent misleading anomalies caused by small sample sizes.

​
Depth Filter

Control how deeply dimensions are combined during analysis.
  • Depth defines the number of dimensions cross-referenced.
  • Lower depth → broader trends.
  • Higher depth → more granular insights (e.g., Region + Device).
Use Depth Filter carefully to avoid excessive dimension combinations and noise in your Top Stories feed.

​
Applying the Preset

1

Configure your filters

Set your KPI, Dimension, Sub-population, and Depth filters based on your analytical goal.
2

Apply Changes

Click Apply Changes to generate Top Stories based on the defined filters.
3

Review results

Top Stories will now reflect only the signals that match your preset — high-priority, high-relevance anomalies.
By combining KPI, Dimension, Sub-population, and Depth filters, you can move from broad monitoring to precise, high-signal anomaly detection. Filter Presets effectively turn autonomous insights into a prioritized and actionable workflow.

​
What to do next

Top Stories Overview

Back to the Top Stories feature overview.

KPI Attribution

Understand contributors, opposers, and side effects once your preset is applied.

Alerts Configuration

Combine Filter Presets with Alerts to get notified only on the signals you care about.