Business Requirements Scoring Guide

Transparent breakdown of how requirements are evaluated and scored

How to Use This Guide

This guide explains the exact criteria and weights used to evaluate your business requirements. Each dimension is scored independently (0-100%) and contributes equally to the final Quality Score.

Overall Scoring Formula

The final Quality Score is the average of all five quality dimensions:

Quality Score = (Clarity + Completeness + Testability + Feasibility + SMART) ÷ 5

Each dimension is weighted equally at 20% of the total score.

Score Interpretation

90-100% Excellent - Minimal improvements needed
70-89% Good - Some minor improvements
50-69% Fair - Significant improvements needed
0-49% Poor - Major rework required

Quick Tips for High Scores

Use specific numbers and metrics
Include clear timeframes
Specify who does what (actors)
Avoid vague subjective terms
Include acceptance criteria
Be concise but complete

Clarity Score

Weight: 20%

Measures how easily the requirement can be understood without ambiguity. Clear requirements minimize misinterpretation.

Criteria	Scoring Impact	Points Range
Clear Actor Identification Explicitly states who performs the action (system, user, admin, etc.)	+15 points	0 or 15 points
Strong Action Verbs Uses definitive verbs: must, shall, will, requires, allows, enables	+15 points	0 or 15 points
Conciseness Optimal length (10-30 words)	+10 points	-10 to +10 points
Vague Phrases Each vague term: fast, user-friendly, intuitive, etc.	-5 points each	0 to -25 points
Passive Voice Using passive instead of active voice	-10 points	0 or -10 points

Base Score: 70 points

Maximum possible addition: +40 points

Maximum possible deduction: -35 points

Effective range: 35-110 → Clamped to 0-100

Good Example (Score: ~90%)

"The payment processing system must validate credit cards within 2 seconds of submission."

Why it scores high: Clear actor (system), strong verb (must), specific metric (2 seconds), no vague terms.

Poor Example (Score: ~40%)

"It should be fast and user-friendly for processing payments."

Why it scores low: No clear actor, weak verb (should), vague terms (fast, user-friendly), passive voice.

Completeness Score

Weight: 20%

Evaluates whether all necessary information is present for implementation. Complete requirements leave no questions unanswered.

Criteria	Scoring Impact	Points Range
Input/Output Specification Clearly defines what goes in and what comes out	+10 points	0 or 10 points
Preconditions States what must be true before the action (when, if, given)	+15 points	0 or 15 points
Post-conditions Defines what happens after (then, as a result)	+15 points	0 or 15 points
Error Handling Specifies how errors/exceptions are handled	+10 points	0 or 10 points
Performance Criteria Includes measurable performance targets	+10 points	0 or 10 points
Missing Essential Info Actor not specified (user, system, admin)	-15 points	0 or -15 points

Base Score: 65 points

Maximum possible addition: +60 points

Maximum possible deduction: -15 points

Effective range: 50-125 → Clamped to 0-100

Good Example (Score: ~85%)

"When a user submits invalid login credentials (input), the system must display an error message (output) within 1 second and log the attempt for security review (post-condition)."

Why it scores high: Clear input/output, precondition (when), post-condition, performance metric (1 second).

Poor Example (Score: ~30%)

"Handle login errors appropriately."

Why it scores low: No input/output, no pre/post conditions, no performance criteria, vague term (appropriately).

Testability Score

Weight: 20%

Assesses how easily the requirement can be verified through testing. Testable requirements have clear pass/fail criteria.

Criteria	Scoring Impact	Points Range
Measurable Criteria Includes numbers, percentages, counts, or time units	+30 points	0 or 30 points
Clear Pass/Fail Conditions Uses definitive terms: must, shall, fail, pass, succeed	+20 points	0 or 20 points
Acceptance Criteria Explicitly mentions verification, validation, or testing	+15 points	0 or 15 points
Subjective Language Each subjective term: intuitive, user-friendly, seamless, etc.	-25 points each	0 to -50 points
Absolute Statements Unrealistic absolutes: always, never, all, every (without metrics)	-20 points	0 or -20 points

Base Score: 60 points

Maximum possible addition: +65 points

Maximum possible deduction: -70 points

Effective range: -10 to 125 → Clamped to 0-100

Good Example (Score: ~95%)

"The search functionality must return results for 95% of queries within 500 milliseconds, as verified by automated performance tests."

Why it scores high: Measurable (95%, 500ms), clear pass/fail (must), acceptance criteria (verified by tests).

Poor Example (Score: ~10%)

"Search should be intuitive and always return relevant results quickly."

Why it scores low: Subjective (intuitive, relevant, quickly), absolute (always), no measurable criteria.

Feasibility Score

Weight: 20%

Evaluates whether the requirement is realistically achievable given technical, resource, and business constraints.

Criteria	Scoring Impact	Points Range
Realistic Expectations Avoids impossible absolutes: always, never, 100%, perfect	No penalty if avoided	0 or -30 points
Conflicting Constraints Combines opposing requirements (e.g., free + maximum security)	-25 points	0 or -25 points
Scope Control Avoids overly broad terms: all, every, entire, complete, full	-10 points	0 or -10 points
Technical Feasibility Uses achievable technology (AI, ML, blockchain recognized as advanced)	+10 points	0 or 10 points
Performance Realism Extremely tight constraints (<10ms) penalized	-20 points	0 or -20 points

Base Score: 75 points

Maximum possible addition: +10 points

Maximum possible deduction: -85 points

Effective range: -10 to 85 → Clamped to 0-100

Good Example (Score: ~80%)

"The reporting module shall generate monthly sales reports for up to 10,000 transactions within 30 minutes during off-peak hours."

Why it scores high: Realistic volume (10,000), reasonable timeframe (30 minutes), constraints defined (off-peak).

Poor Example (Score: ~25%)

"The system must process all transactions instantly with 100% accuracy and zero cost."

Why it scores low: Impossible absolutes (all, instantly, 100%, zero), conflicting constraints (100% accuracy + zero cost).

SMART Criteria Score

Weight: 20%

Evaluates how well the requirement follows the SMART framework: Specific, Measurable, Achievable, Relevant, Time-bound.

SMART Component	Scoring Criteria	Points Range (per component)
Specific (20 points each) Clear actor, action, scope, reason, constraints	5 criteria × 20 points each	0-100 points
Measurable (25 points each) Quantifiable metric, verification method, success criteria, baseline	4 criteria × 25 points each	0-100 points
Achievable Technical feasibility, resource availability, organizational fit, compliance	Base 50 ± adjustments	0-100 points
Relevant Business goal link, stakeholder need, strategic fit, priority justification	Base 50 ± adjustments	0-100 points
Time-bound Fixed deadline (50), relative timeframe (30), milestone (20), none (0)	Single selection	0, 20, 30, or 50 points

Calculation: (S + M + A + R + T) ÷ 5

Specific (S): 0-100 points

Measurable (M): 0-100 points

Achievable (A): 0-100 points

Relevant (R): 0-100 points

Time-bound (T): 0-100 points

Overall SMART Score: Average of all 5 components

SMART Example (Score: ~91%)

"The checkout system must complete 90% of transactions in under 5 seconds by Q3 2024 to improve conversion rates by 15%."

SMART Breakdown: Specific (what, why), Measurable (90%, 5s, 15%), Achievable (realistic), Relevant (conversion), Time-bound (Q3 2024).

Non-SMART Example (Score: ~26%)

"Make the checkout process better soon."

Why it fails: Not specific (better?), not measurable, no timeframe (soon), relevance unclear, achievability unknown.

How Real-time Validation Works

The real-time validation feature analyses your requirements as you type, providing immediate feedback on common quality issues. Here's how it works:

Text Analysis

As you type in the requirement textarea, the system continuously analyses your text for:

Check Type	What it looks for	Frequency
Vague Phrases	Words like "fast", "user-friendly", "intuitive", "soon"	Every keystroke
Missing Elements	Actor (who), action (what), metrics (how much), timeframe (when)	Every keystroke
Passive Voice	"is processed" vs. "the system processes"	Every keystroke
Subjective Language	"good", "bad", "nice", "excellent"	Every keystroke

Instant Detection

The system uses regular expressions and pattern matching to identify issues immediately. For example:

Input: "The system should be fast and user-friendly"

Detected: 2 vague phrases ("fast", "user-friendly")

Input: "The system must process orders within 2 seconds"

Status: Good - Clear actor, action, and measurable timeframe

Visual Feedback

Based on the number and severity of issues detected, the status indicator changes:

Green (Good)
0-1 minor issues

Orange (Warning)
2-3 issues detected

Red (Problem)
4+ significant issues

Gray (Ready)
No text entered

The status dot also includes subtle animations:

Green: Solid - no issues detected
Orange: Gentle pulse - minor issues that should be addressed
Red: Faster pulse - significant issues requiring attention

Specific Issue Listing

When issues are detected, they're listed in the "Issues Detected" section with specific feedback:

Example Issues List:

1. Vague phrase: "fast" - Use concrete metrics instead
2. Missing measurable criteria - Add numbers or percentages
3. Unclear actor - Specify who performs this action

The system shows up to 5 issues at a time to avoid overwhelming the user. If more than 5 issues are detected, it shows "... and X more issues."

Performance & Behavior

The real-time validation is designed to be:

Fast: Analysis happens within milliseconds of each keystroke
Lightweight: Uses minimal resources to avoid slowing down typing
Non-blocking: Doesn't interrupt your typing flow
Educational: Provides specific guidance, not just criticism

Note: Real-time validation is a preliminary check designed to catch common issues. The full "Analyse Requirement" button performs a more comprehensive analysis that includes scoring across all dimensions and checks for redundancy with existing requirements.

Try the Scoring System

Enter a requirement below to see how it would be scored across all dimensions:

Scoring Analysis

Important Notes

All scores are clamped between 0% and 100% after calculation
Vague phrases are cumulative (multiple instances increase penalty)
The scoring algorithm is rule-based and automated - human review is still recommended
Requirements with scores below 60% typically need significant rework
This is a simulation for guidance - actual requirement analysis may require domain expertise