The CASTLE framework: Turning Data Into Decisions for Product Teams

Let’s be honest: product metrics can be a bloody mess. You’re drowning in numbers, unsure which ones matter, and it feels like you’re playing detective with user behaviour. This is where the CASTLE Framework, developed by Nielsen Norman Group, comes in—a structured and practical approach to measuring and improving user experience (UX).

I heard about it by the creator at some online UX thing a year or two ago ??♀? and was taken by how it really took a structured approach to making sense of key UX and CX data to improve products - and is heavily inspired by several old school Business analysis approaches to sense-make and communicate data and recommendations… which this old-school ex BA is a massive fan of.

What is the CASTLE Framework?

The CASTLE Framework is a human-centred metrics model developed by Nielsen Norman Group to evaluate and improve digital experiences through six dimensions:

CASTLE stands for:

• Cognitive Load: How much mental effort is required to complete a task?
• Advanced Feature Usage: Are users taking full advantage of the product’s key features?
• Satisfaction: How happy are users with their overall experience?
• Task Efficiency: How quickly and effectively can users complete tasks?
• Learnability: How intuitive is the system for first-time users?
• Errors: How often do users make mistakes, and why?

These six dimensions provide a roadmap for identifying where products fall short, guiding both strategic actions and tactical fixes. Whether designing new products or refining existing ones, CASTLE keeps product teams focused on what matters most: creating seamless, intuitive, and delightful experiences.

How CASTLE Enhances the Product Lifecycle

The CASTLE Framework isn’t just a tool for post-launch evaluation—it can be applied throughout the product lifecycle. Here’s how:

• During Discovery and Research: Identify high-friction tasks and unmet user needs by measuring cognitive load and satisfaction in current systems.
• In Design and Development: Use learnability and error metrics to guide prototyping and usability testing, ensuring a smoother experience.
• At Launch: Track advanced feature usage and task efficiency to validate product goals and identify areas for iteration.
• For Continuous Improvement: Use satisfaction and error metrics to uncover opportunities for enhancing existing features or retiring underused ones.

How is CASTLE Different from the HEART Framework?

If you’re familiar with Google’s HEART Framework, you might wonder how CASTLE compares. Both frameworks aim to evaluate user experience, but they serve slightly different purposes:

HEART (Happiness, Engagement, Adoption, Retention, Task Success): HEART is broader and comprises key user engagement and satisfaction indicators over the product lifecycle. And that’s particularly helpful when you’re evaluating long-term trends like retention and adoption.

CASTLE (Cognitive Load, Advanced Feature Usage, Satisfaction, Task Efficiency, Learnability, Errors): CASTLE is all about understanding how the product works and what users experience as pain points when using a product. It’s more fine-grained, giving you insights to tweak user flows and system design.

In short, HEART is ideal for strategic, overarching product goals, while CASTLE excels at tactical, hands-on UX improvements. Together, they complement each other, with HEART guiding what to measure and CASTLE focusing on how to dig into the details. Key differences include:

• Focus: CASTLE emphasises cognitive load, learnability, and errors—factors that impact usability at a detailed level.
• Application: HEART is more suited to tracking high-level user experience, while CASTLE excels at guiding iterative improvements during the product lifecycle.
• Data Source: CASTLE integrates operational data, product analytics, and qualitative insights.

?

This Is Like the SUS Score-Is It Same?

Another popular usability measure is the System Usability Scale (SUS), but it’s not the same as CASTLE. Here’s how they differ:

SUS: A 10-question survey to get an immediate quantitative snapshot of a system’s usability. You’re awarded one score that essentially sums up the whole user experience but no more specifics about whether or not the system is useful for the user.

CASTLE: A multi-dimensional assessment of user experience, which takes into account specifics like cognitive load and learnability. CASTLE doesn’t just measure a score, it also identifies key areas for growth.

Basically think of SUS as the "usability pulse check" — simple, fast, and useful for benchmarking. CASTLE, however, is more of a diagnostics package, giving you the broken down information you need to identify specific issues and improve your product, SUS is handy for a broad usability score, but not enough of the useful and relevant information that CASTLE provides.

CASTLE extends beyond the simplicity of SUS by providing:

• Broader Scope: CASTLE covers six distinct dimensions, not just overall usability.
• Actionable Insights: Unlike SUS, CASTLE identifies specific areas for improvement, such as task efficiency or error reduction.
• Lifecycle Integration: CASTLE can be applied at every stage of the product lifecycle, from design to continuous improvement.

How to start using the CASTLE Framework:

The CASTLE Framework isn’t just about understanding what’s happening with your product—it’s about turning those insights into actions. To help product teams visualise how to apply the framework, here are two examples showcasing how call centre metrics, product analytics, and operational data can be mapped against CASTLE.

?

Understanding the Building Blocks of CASTLE

The CASTLE Framework is all about turning data into actionable insights. To make this framework even more practical for product teams, I’ve adapted it slightly from the original Nielsen Norman Group (NNG) version. My approach adds Insights and Next Steps columns, making it easier (in my opinion) to move from analysis to action.

This adaptation is particularly helpful when presenting findings and recommendations to stakeholders—it bridges the gap between data and decisions.

Here’s a quick breakdown of the columns in the framework:

• Goals The ultimate objectives tied to your product strategy or user experience outcomes. Goals should be specific, measurable, and aligned with both user and business needs. Example: "Reduce friction during onboarding to improve user retention."
• Signals Observable behaviours or patterns that indicate whether you’re moving closer to—or further from—your goal. Signals help you identify where to focus your analysis. Example: "High drop-off rates at Step 2 in the registration process."
• Measures The quantifiable metrics used to validate signals and assess performance. Measures should be specific and data-driven. Example: "Drop-off percentage at each step of the onboarding flow."
• Insights The actionable understanding you derive from analysing signals and measures. Insights translate data into a story that highlights the root cause of problems or opportunities. Example: "Users are abandoning onboarding at Step 2 because the form requires too much information upfront."
• Next Steps The concrete actions to address insights and improve outcomes. These are essential for driving data-informed decisions and implementing meaningful change. Example: "Redesign Step 2 to prioritise essential information and defer non-critical inputs to later stages.

Examples of CASTLE in action:

To bring the CASTLE framework to life, I’ve put together some simplified examples inspired by initiatives I’ve worked on. These examples showcase how the framework can be applied in real-world scenarios, with a few of my personal “hacks” to enhance its impact.

For each example, I’ve included a customer scenario and a product team perspective to provide context and show how this framework supports both sides of the experience.

Let’s dive in and explore how CASTLE can drive meaningful outcomes for customers and teams alike. ??

Example 1: Improving Onboarding for a Subscription Service

Customer Scenario: Emma tries signing up for a streaming service but gets stuck on Step 2 of the onboarding process, which asks for multiple details upfront. She abandons the process out of frustration and opts for a competitor’s service.

Product Team Scenario: The team notices a high drop-off rate during Step 2 and sees customer complaints about the complexity of onboarding. They redesign the flow, deferring non-essential details to later, and notice a 20% increase in successful signups.

Example 2: Optimising Checkout for an E-Commerce Platform

?Customer Scenario: Liam tries to purchase a laptop but abandons the checkout after encountering unclear payment options and additional fees that weren’t disclosed upfront.

Product Team Scenario: The team identifies that 30% of drop-offs occur during the payment stage due to a cluttered interface. They streamline the payment options, resulting in a smoother user experience and higher conversion rates.

How This Framework Supports Your Product Team

These tables aren’t just a summary of issues—they’re a roadmap for action. By breaking down the signals (what’s happening), measures (how to track it), and key insights (what’s wrong), teams can define next steps that align with clear goals. Whether you’re designing a new feature or improving an existing process, the CASTLE Framework ensures every metric is purposeful and actionable.

How CASTLE Supports Strategic Goals

The real power of CASTLE lies in its ability to link metrics to broader product and business goals:

• Enhancing Customer Experience (CX): Cognitive load and satisfaction metrics identify friction points, helping teams prioritise improvements that directly impact user happiness.
• Improving Employee Experience (EX): Errors and task efficiency metrics reduce repetitive tasks for frontline teams, boosting morale and productivity.
• Achieving Omnichannel Success: Advanced feature usage and learnability metrics guide seamless integration across platforms and devices.
• Driving Data-Driven Decisions: CASTLE metrics provide actionable insights that validate product strategies and inform the roadmap.

Final Thoughts: Why I think CASTLE will benefit you and your Product Teams

Metrics are more than numbers—they’re your map to understanding users, solving problems, and delivering exceptional experiences. CASTLE transforms data into decisions, helping product teams create seamless, intuitive, and delightful digital journeys.

By integrating customer and product team scenarios, CASTLE connects the dots between user experience, operational data, and strategic goals. Whether you’re designing new features or fixing old pain points, CASTLE ensures every action is purposeful and impactful.

Resources to Explore Further

1. Nielsen Norman Group: The CASTLE Framework Explore the origin and application of the CASTLE Framework, designed for improving UX and product performance. URL: https://www.nngroup.com/articles/castle-framework/
2. Google Research: The HEART Framework A complementary framework to CASTLE, focusing on Happiness, Engagement, Adoption, Retention, and Task Success. URL: https://research.google/pubs/pub36299/
3. UX Design: A Beginner’s Guide to UX Metrics This guide covers essential UX metrics, including practical tips for improving user experience. URL: https://uxdesign.cc/beginners-guide-to-ux-metrics-4b9a7a5cfe8a
4. Mixpanel Blog: How to Integrate Metrics into Product Roadmaps Learn how to incorporate data-driven insights into your product roadmap to maximise impact. URL: https://mixpanel.com/blog/integrate-metrics-into-product-roadmap/
5. Measuring Usability: System Usability Scale (SUS) A practical guide to the SUS score, its use, and how it compares to frameworks like CASTLE. URL: https://measuringu.com/sus/