AgentSkillsCN

ux-task-estimator

估算用户工作流所需的时间、点击次数以及各项效率指标。在衡量任务完成效率、对比不同设计方案、设定用户体验基准,或当用户提到“任务耗时”、“点击次数”、“效率”、“任务完成度”、“基准”或“所需时间”时,可加以应用。

SKILL.md
--- frontmatter
name: ux-task-estimator
description: Estimate time-on-task, click counts, and efficiency metrics for user workflows. Use when measuring task completion efficiency, comparing design alternatives, setting UX benchmarks, or when user mentions "time on task", "click count", "efficiency", "task completion", "benchmark", or "how long does it take".

UX Task Estimator

Estimate and measure task completion metrics for UX optimization.

Core Metrics

Time on Task (ToT)

Estimated time from task start to completion.

Components:

  • Reading/scanning time
  • Decision time
  • Input time
  • System response wait time
  • Navigation time

Click/Tap Count

Total interactions required to complete task.

Types:

  • Navigation clicks
  • Selection clicks
  • Action clicks (submit, confirm)
  • Corrective clicks (back, undo)

Error Rate Estimate

Likelihood of user making mistakes.

Factors:

  • Input field complexity
  • Ambiguous options
  • Lack of validation
  • Confusing labels

Estimation Framework

Reading Time

Average reading speed: 200-250 words per minute

Content TypeTime per Item
Short label0.5 sec
Button text0.5 sec
Sentence2-3 sec
Paragraph10-15 sec
Full form instructions15-30 sec

Decision Time (Hick's Law)

Base time + (150ms × log2(options + 1))

OptionsAdded Time
2~300ms
4~450ms
8~600ms
16~750ms

Input Time

Input TypeTime
Click/tap0.2 sec
Simple text (5 chars)2 sec
Email address4 sec
Password3 sec
Dropdown selection1.5 sec
Checkbox0.5 sec
Date picker3 sec

Navigation Time

ActionTime
Page load wait1-3 sec
Visual scan for target1-2 sec
Mouse/finger movement0.3-0.5 sec
Scroll action0.5 sec

Task Analysis Process

  1. List all steps in the task flow
  2. Categorize each step (read, decide, input, navigate)
  3. Apply time estimates per step
  4. Sum total time and count interactions
  5. Add buffer for errors/confusion (10-20%)

Output Format

code
# Task Analysis: [Task Name]

**Goal:** [User's objective]
**Starting Point:** [Where user begins]
**End Point:** [Task completion criteria]

## Step-by-Step Breakdown

| Step | Action | Type | Time | Clicks |
|------|--------|------|------|--------|
| 1 | Read page title | Read | 0.5s | 0 |
| 2 | Scan navigation | Read | 1.5s | 0 |
| 3 | Click "APIs" menu | Navigate | 0.5s | 1 |
| 4 | Wait for page load | Wait | 2s | 0 |
| 5 | Find target API | Read/Decide | 2s | 0 |
| 6 | Click API name | Navigate | 0.5s | 1 |
| ... | ... | ... | ... | ... |

## Summary

| Metric | Value | Benchmark | Status |
|--------|-------|-----------|--------|
| Total time | Xs | <Ys | ✅/⚠️/❌ |
| Click count | X | <Y | ✅/⚠️/❌ |
| Decision points | X | <Y | ✅/⚠️/❌ |
| Error risk points | X | 0 | ✅/⚠️/❌ |

## Optimization Opportunities

| Step | Current | Optimized | Savings |
|------|---------|-----------|---------|
| [Step] | Xs/Xclicks | Ys/Yclicks | Z% |

## Recommendations
1. **[Change]**: Reduces [metric] by [amount]

Benchmarks by Task Type

Simple Action (view, read)

  • Target: <10 seconds, <3 clicks
  • Example: View dashboard stats

Standard Action (create, edit)

  • Target: <30 seconds, <10 clicks
  • Example: Create new item

Complex Action (configure, setup)

  • Target: <2 minutes, <20 clicks
  • Example: Configure integration

Wizard/Flow (multi-step)

  • Target: <5 minutes, <30 clicks
  • Example: Complete onboarding

Efficiency Ratios

Clicks per Goal

Total clicks ÷ number of tasks completed

Good: <5 clicks per task Acceptable: 5-10 clicks Poor: >10 clicks

Time per Click

Total time ÷ total clicks

Good: <2 seconds (user flows smoothly) Acceptable: 2-5 seconds (some thinking) Poor: >5 seconds (confusion/waiting)

Comparison Template

code
# Task Comparison: [Task Name]

| Metric | Current Design | Proposed Design | Improvement |
|--------|---------------|-----------------|-------------|
| Time on task | Xs | Ys | Z% faster |
| Click count | X | Y | Z fewer |
| Error risk | X points | Y points | Z% safer |
| Pages visited | X | Y | Z fewer |

References

For detailed estimation models: See references/estimation-models.md