Advanced Concepts
Outline
Welcome
In this module, you’ll learn how to unlock deeper insights and improve your data storytelling in Optimizely Analytics using advanced features like sparklines, sampling, and period-over-period comparisons. These tools help you analyze large datasets more efficiently, highlight meaningful trends, and compare performance across different time periods—all while maintaining clarity and precision in your visualizations.
Whether you're building dashboards or exploring ad hoc reports, these capabilities will help you create more dynamic and actionable analyses.
After completing this module, you’ll be able to:
Create sparklines to visualize trends across multiple metrics in a compact format
Apply sampling to improve query performance and understand its impact on accuracy
Enable period-over-period comparisons in funnels, event segmentation, and retention analyses to track changes overtime
Create Sparklines
Sparklines are compact, simplified charts designed to show trends over time across multiple measures—without taking up much space. They’re ideal for dashboards and quick comparisons where you want to show movement rather than exact values.
In Analytics, you can create sparklines to easily visualize metric patterns side-by-side. These visual summaries help you compare fluctuations, detect anomalies, or highlight progress across key business metrics.
Learn how to create sparklines by visiting the documentation: Create Sparklines
What Is Period-over-Period Comparison?
Period-over-period comparison lets you evaluate how your current performance compares to a past time frame. This feature is available in several types of explorations, providing visibility into metric trends and changes.
In Funnels
The Show Over Time feature in funnel explorations supports period comparisons for:
% converted between specific stages
Time between specific stages
Converted to specific stage
Learn how to enable period-over-period comparison in funnels: https://optimizely.navattic.com/exh0yx9
In Event Segmentation
All measures in Event Segmentation support period-over-period comparisons. This allows you to see how event trends evolve across selected timeframes.
Learn how to enable this in Event Segmentation: https://optimizely.navattic.com/doh01ym
In Retention Analysis
Retention analysis also supports over-time comparisons across both retention measures, giving you a historical view of user return behavior.
Learn how to enable this in Retention analysis: https://optimizely.navattic.com/sip06zz
What Is Sampling?
Sampling is a technique used to improve query performance by analyzing a statistically representative portion of your dataset instead of the entire volume of data. This allows faster, yet still highly accurate, results—especially when dealing with large datasets.
Analytics offers two sampling modes:
Faster Response (default):Uses a lower sampling rate to return faster results, with slightly lower precision.
Higher Precision:Uses a higher sampling rate to deliver more accurate results, with a tradeoff in performance speed.
Sampling in Explorations
When sampling is applied in an exploration, each data point in the visualization includes a 95% error bound, displayed in the tooltip.
For example:
If a value is 100 and the error bound is 1%, the unsampled value likely falls between 99 and 101.
To improve readability:
Red dots are used to highlight data points with >5% error.
A sampled indicator can toggle these highlights on or off.
Learn how to set up sampling within an exploration: https://optimizely.navattic.com/a2510rnj .
Note: Sampling is a feature configured at the application level. Contact Optimizely Support to enable sampling for your application.
Learn more about Sampling in this documentation: Sampling