Skip to main content

​
Overview

Beyond monitoring current metrics, the Dashboard provides powerful analytical capabilities to explore historical data, identify patterns, and derive actionable insights. This guide introduces key analysis techniques and tools.

​
Analysis capabilities

Transform raw metrics into meaningful insights:

Trend analysis

Identify long-term patterns, seasonal variations, and growth trajectories.

Comparative analysis

Compare metrics across time periods, segments, or dimensions.

Correlation discovery

Find relationships between different metrics and variables.

Anomaly detection

Automatically identify unusual patterns and outliers in your data.

​
What you will learn

This exploration task teaches you to:
  • Adjust time ranges for different analysis perspectives
  • Use filters and segments to drill down into data
  • Apply comparison modes to benchmark performance
  • Export data for external analysis
  • Create and save analytical views

​
Prerequisites

  • Completed Setting up dashboard alerts
  • Dashboard with historical data (at least a few days of data)
  • Basic understanding of the metrics you want to analyze
1

Select appropriate time range

Choose a time period that provides meaningful context:
  • Day view: For recent changes and hourly patterns
  • Week view: For weekly cycles and short-term trends
  • Month view: For monthly patterns and medium-term trends
  • Custom range: For specific events or comparisons
Longer time ranges reveal trends; shorter ranges show immediate changes.
2

Apply data filters

Narrow your focus using available filters:
  • Geographic regions
  • User segments or customer types
  • Product categories or service tiers
  • Time of day or day of week
Start with broad filters and progressively narrow to isolate specific patterns.
3

Use comparison mode

Enable comparison features to benchmark performance:
  • Compare to previous period (day over day, week over week)
  • Compare to same period last year for seasonality
  • Compare across segments (mobile vs desktop, region A vs region B)
Comparison lines and percentage changes help identify improvements or degradations.
4

Drill down into details

Click on chart data points to explore deeper:
  • View detailed records behind aggregated metrics
  • See breakdowns by sub-categories
  • Export specific data segments
Drilling down may reveal sensitive individual records. Ensure you have permission to view detailed data.
5

Identify correlations

Look for relationships between different metrics:
  • Do traffic spikes correlate with conversion drops?
  • Does response time affect user engagement?
  • Are there lag effects between marketing spend and revenue?
Correlation does not imply causation. Use correlation as a starting point for deeper investigation.
6

Annotate significant events

Add markers to charts for important events:
  • Product launches or updates
  • Marketing campaigns
  • Infrastructure changes
  • External events (holidays, news)
This helps explain metric changes and builds organizational knowledge.
7

Export data for deeper analysis

When dashboard tools are insufficient:
  • Export data to CSV for spreadsheet analysis
  • Use API access for programmatic analysis
  • Download charts for presentations and reports

​
Analytical techniques

​
Moving averages

Smooth out noise to see underlying trends:
  • 7-day average: Reduces daily fluctuations
  • 30-day average: Shows monthly trends
  • Apply to volatile metrics like daily active users or revenue

​
Percentile analysis

Understand distribution rather than just averages:
  • Median (50th percentile): Typical user experience
  • 95th percentile: Experience of worst-affected users
  • 99th percentile: Edge cases and outliers

​
Cohort analysis

Track groups of users over time:
  • Group users by when they first engaged
  • Compare behavior across cohorts
  • Identify retention patterns and lifecycle changes

​
Common analysis scenarios

  1. Check the time range of the change
  2. Look for correlated changes in other metrics
  3. Review annotations for relevant events
  4. Filter by segments to see if change is universal or isolated
  5. Compare to historical patterns for similar events
  1. Identify metrics that correlate with Y over time
  2. Use lead/lag analysis to find predictive indicators
  3. Segment data to find which groups contribute most
  4. Test hypotheses by filtering different dimensions
  1. Compare short-term and long-term trend lines
  2. Look for seasonal patterns that may repeat
  3. Analyze underlying drivers for the trend
  4. Consider external factors that might change

​
Creating analytical reports

Save your analysis for future reference:
  1. Configure your dashboard with the right time range and filters
  2. Add explanatory text widgets with your findings
  3. Save as a named view: “Q1 Analysis - Traffic Sources”
  4. Share with stakeholders or schedule automated delivery

​
Next steps

Now that you can analyze data effectively, learn about Generating dashboard reports to share insights with your team and stakeholders.