Metrics

Overview

Metrics helps you understand how a completed evaluation performed and where a benchmark is improving or struggling over time.

This page currently shows analytics for Evaluations. Use it to:

• Review the results of a single completed evaluation.
• Compare multiple completed evaluations within the same benchmark.
• Spot stronger and weaker guideline categories.
• Identify which conditions caused a lower score.
• Document findings with notes and screenshots.

The page has two sections:

• Overview a visual summary of performance
• Details a full question-by-question breakdown

The Overview Section

The Overview tab gives you a quick read on the latest evaluation and how that benchmark is trending over time.

What Each Card Shows

• Latest Score: The overall result for the most recent evaluation of the selected benchmark.
• Guideline Count: How many criteria exist in each guideline category.
• Trend: Whether results for the same benchmark are improving, declining, or staying flat over time.
• Category Breakdown: How each category scored based on pass and fail results.
• Category Trends: How each category changed across multiple evaluation sessions for the same benchmark.
• Category Radar: The same category performance shown in a radar view for a different way to compare results.

UXit shows the same evaluation data in multiple visual formats so users can scan it in the way that works best for them. Some people want a quick score, some want to compare categories, and some want to look for patterns over time.

Reading The Charts

• Use Latest Score and Trend for a quick summary.
• Use Category Breakdown to see which categories are strongest or weakest right now.
• Use Category Trends to see whether a category is improving, declining, or staying stable across versions.
• Use Category Radar when you want a visual comparison of overall category shape and balance.

In the radar chart, a larger and more evenly filled shape generally indicates stronger overall category performance.

Customize The View

Use Customize to show or hide chart legends.

Hiding legends can reduce visual clutter and free up space inside each card.

Even when legends are hidden, most interactive charts still reveal category names when you hover over the chart data.

The Details Section

The Details tab lives on the same analytics screen in the product, but it is documented separately here so the charting view and the question-level breakdown are easier to explain.

Use Metrics to understand the summary and trend views in the Overview tab, then see Condition Details for the full searchable table and per-condition notes and image references.

Comparing Evaluations Over Time

Metrics becomes most useful when the same benchmark has been evaluated more than once with the same guideline set.

This allows you to:

• Compare multiple evaluations under the same benchmark.
• See whether changes improved the outcome.
• Catch regressions after updates.
• Track category-level progress across evaluation sessions.

If the benchmark stays the same but the design changes, add another evaluation to that benchmark so the trend data remains useful. If you change the guideline set, comparisons may still help, but they become less reliable.

For a concrete example of how to interpret score changes across runs, see Reading Results.

Best Practices

• Keep the benchmark stable: Compare runs under the same benchmark when the core experience area has not changed.
• Keep guidelines consistent: Reusing the same guideline set makes trends easier to trust.
• Start in Overview: Use the charts to spot weak areas quickly before digging deeper.
• Confirm in Condition Details: Use question-level results to find the exact failures behind a lower score.
• Document what you find: Add notes and screenshots so the next evaluation has useful context.
• Review Methodology when needed: If you need the scoring logic behind these visuals, see the Methodology page.