Build a UX Metrics Dashboard

Introduction

This tutorial shows how to use the Results API QXscore endpoint to build a UX metrics dashboard that tracks experience quality across tests and time.

QXscore is a composite benchmark metric that combines behavioral and attitudinal signals, making it ideal for dashboards, trend analysis, and executive reporting.

What you’ll build

A dashboard-ready data flow that:

1. Retrieves QXscores for a test
2. Understands QXscore structure and components
3. Normalizes QXscore data for BI tools
4. Enables trend and comparison views across tests

Target audience

• Product managers
• UX researchers
• Analytics and BI engineers
• Research operations teams

Prerequisites

• A valid access token (ACCESS_TOKEN). Go to Authorization for details.
• A known testId. Go to How to obtain a TestId (UUID) for details.
• A BI or visualization tool (Tableau, Looker, Power BI, etc.)

What is QXscore?

A QXscore is a standardized metric (0–100) that evaluates experience quality by combining:

Behavioral component

• Derived from navigational task success
• Measures what participants did

Attitudinal component

• Derived from questionnaire responses
• Measures what participants felt
• Divided into four categories:
  • Usability
  • Trust
  • Appearance
  • Loyalty

Each test can contain one or more QX task groups, each producing a QXscore. Task groups contain at least two tasks each.

📘

The QXscore endpoint returns aggregated data only. This means that returned scores are an average of all participant sessions .

Steps

Step 1 — Retrieve QXscores for a test

Endpoint

What this returns

• One or more QX task groups
• Aggregate scores across completed sessions
• Component-level and subcomponent-level values

Example (curl)

curl --location 'https://api.use2.usertesting.com/api/v2/testResults/TEST_ID/qxScores' \
  --header 'Authorization: Bearer ACCESS_TOKEN' \
  --header 'Content-Type: application/json'

Step 2 — Understand the response structure

Top-level fields

{
  "testId": "TEST_ID",
  "qxScores": [...],
  "meta": {
    "totalQxTasks": 1,
    "completes": 12
  }
}

• testId: indicates the test ID
• meta.completes: indicates the number of completed sessions used in scoring

QX task group structure

Each entry in qxScores[] represents one QX task group.

{
  "taskGroupId": "TASK_GROUP_ID",
  "label": "QXscore 1",
  "qxScore": 38,
  "components": {
    "behavioral": 25,
    "attitudinal": 50
  },
  "values": {
    "behavioral": [50, 50],
    "usability": 50,
    "trust": 50,
    "appearance": 50,
    "loyalty": 50
  }
}

• qxScore: the task group's QXscore, calculated by the average of the behavioral and attitudinal values found in the components property.
• components.behavioral: indicates the average obtained from the navigational task scores in values.behavioral, which display the session average for each task
• components.attitudinal: indicates the average obtained from values.usability, values.trust, values.appearance, and values.loyalty, which indicate the session average for each questionnaire category

Step 3 — Model QXscore data for analytics

Recommended warehouse table: qx_scores

Step 4 — Create dashboard views

Once loaded into a BI tool, QXscore enables several high-value UX views.

1. Overall UX health

   • Metric tile showing latest QXscore
   • Compared against internal benchmarks

2. Behavioral vs attitudinal split

   • Stacked bar or dual-axis chart
   • Identifies execution vs perception gaps

3. Subcomponent breakdown

   • Radar or bar chart for:
     • Usability
     • Trust
     • Appearance
     • Loyalty

4. Longitudinal trends

   • QXscore by test launch date
   • Track improvement over time

Step 5 — Compare across tests or releases

By storing QXscores per test, teams can:

• Compare redesign A vs redesign B
• Track UX impact across releases
• Correlate QXscore with:
  • Conversion
  • Retention
  • NPS
  • Support tickets

Error handling

Take into consideration the following:

• For scheduling, consider placing automated retries and notifications if you encounter a 503 error.
• Consider placing backoff logic on 429 error responses.

Common pitfalls

What you can build next

With QXscore data in place, teams can:

• Combine UX and product analytics in BI tools
• Power executive dashboards with experience metrics
• Feed scores into AI systems for automated insights
• Detect UX regressions early

Summary

You now have a clean, scalable way to:

• Retrieve standardized UX quality metrics
• Store them in analytics-friendly formats
• Visualize experience quality across tests and time

QXscore dashboards turn qualitative research into quantifiable, trackable UX signals that organizations can act on with confidence.