Skip to main content

Overview

Smart Analysis is SkoutLab’s core capability — transforming raw data into actionable insights through AI-powered hypothesis generation and validation.

How It’s Different

Traditional BISkoutLab
You build dashboardsAI explores for you
You write queriesYou ask questions
Charts to interpretFindings explained
Manual explorationAutomated discovery

Analysis Modes

Question-Driven

Ask a specific question and get a focused analysis:
“Why did customer churn increase in Q4?”
SkoutLab will:
  1. Identify relevant data
  2. Generate hypotheses about causes
  3. Test each hypothesis
  4. Report validated findings

Exploratory

Don’t know what to ask? Let SkoutLab explore:
“What’s interesting in our sales data?”
The AI will:
  1. Scan for patterns and anomalies
  2. Identify significant trends
  3. Surface unexpected relationships
  4. Highlight areas needing attention

The Analysis Process

SkoutLab uses Knowledge Builder’s learnings to understand:
  • What metrics matter
  • How tables relate
  • What’s “normal” for your data
Based on your question (or general exploration), AI generates testable hypotheses:
  • “Revenue dropped because of Product X”
  • “Churn correlates with support ticket volume”
  • “Sales peak on Tuesdays”
Each hypothesis is rigorously tested:
  • Significance testing
  • Effect size calculation
  • Confidence intervals
Results are compiled into a report:
  • Key findings ranked by impact
  • Evidence for each finding
  • Recommended next steps

Analysis Settings

When creating a task, you can configure:
SettingDescription
Data SourcesWhich connections to analyze
QuestionYour specific question (optional)
DepthHow many hypotheses to generate

Tips for Better Results

Be Specific

Include time ranges, segments, and metrics of interest

Start Broad

If unsure, start with exploratory analysis, then drill down

Iterate

Use findings to formulate more specific follow-up questions

Connect Data

More data sources enable richer cross-cutting insights