Metrics Planning

46 prog-metrics-planning

Metrics Planning helps to identify how and why your customers use your products and services the way that they do. Understanding how your customers engage is essential in tailoring interactions and maximizing conversions.

Explore More.

Related Mindset:

Value-Driven

Data-Driven

Segment:

Program

Inputs:

Validated user stories

Outputs:

User stories updated with necessary acceptance criteria around metrics gathering as well as additional user stories for testing and validation tasks

Metrics planning is the process which enables us to match available attributes and properties at each touch point with a desired outcome or conversion.

For example, in order to increase sales of surfboards, we may determine that personalizing content, or recommending products, based on certain geographic locations is more likely to meet a customer’s needs.

To understand customer segmentation, there are a number of common metrics that can provide value. These may include, but are not limited to:

  • Geographic location: Can be inferred from network location, or collected with permission based on device location.
  • Referer: Can be identified by search engine terms, external websites addresses, campaigns, advertisements or newsletters.
  • Browsing history: When allowed, tracking cookies or other unique identifiers can help to identify browsing trends, or on-site browsing history can be easily inferred.
  • Profile information: With a connected social account or known account information, we can use specific personally identifiable information to determine preferences.
  • Duration: The amount of time that a user spends on a specific page, or overall session duration.
  • Page views: Each individual page request from a browser.
  • Events: Custom actions tracked on a web page or from an application, such as clicking a specific button, or expanding an accordion.

In order to validate and optimize the personalization of content for a particular segment, it is important to test each interaction through collected analytics against the desired conversion.

  • A/B or multivariate testing: With a large dataset, we can determine segmentation profile preferences for multiple alternative content options.
  • Heatmaps: Mouse or trackpad movements, or touch-based tracking can help to identify the most popular areas of a particular page that a user might interact with.

Common Pitfalls

  • Don’t stop testing: Analytics become more valuable as you accumulate more data over a longer period of time. Statistical anomalies are more easily identified and variations in trends over time allow you to adjust certain strategies as they become more or less popular over time.
  • Tailor the testing to the content, not the other way around: Experience is still number one, don’t sacrifice the quality of your content or interactions for metrics unnecessarily.
  • Test everything, but optimize one thing at a time: Changing too many variables at once can lead to unreliable results and false positives.
  • Multivariate testing is more valuable than period-based testing: Trends change over time, comparing results before and after a change may falsely correlate to a content change, rather than simply different seasons or world events.

Tools

The following tools are useful in planning for metrics, gathering metrics, or performing real-time user testing: