Digital Karma Scoring Methodology v2.1
Transparent, automated quality assessment for AI-native websites.
What is Digital Karma?
Digital Karma is a composite quality score (0.00 to 1.00) that measures how well a website implements AI-native principles. Unlike traditional SEO metrics that can be gamed or subjective reputation systems, Digital Karma:
- ✅ Is calculated automatically from objective technical signals
- ✅ Uses transparent formulas — no secret sauce or black boxes
- ✅ Measures real quality — endpoint presence, data freshness, technical excellence
- ✅ Can't be easily gamed — requires genuine implementation
- ✅ Updates regularly — reflects current site status, not past glory
The 7-Signal Formula
Digital Karma Score is calculated from seven weighted quality signals:
Total Score = (Schema Coverage × 0.20) + (Content Freshness × 0.15) + (AI Endpoints × 0.25) + (Federation Presence × 0.15) + (External Links × 0.10) + (Technical Quality × 0.10) + (Dataset Quality × 0.05)
Signal 1: Schema Coverage (20% weight)
What it measures: Quality and completeness of Schema.org structured data markup.
Scoring criteria:
- 1.0: Organization, WebSite, WebPage entities present with all recommended properties
- 0.8: Main entities present, some optional properties missing
- 0.6: Basic entities present, many properties missing
- 0.4: Minimal Schema.org markup
- 0.0: No Schema.org markup
Why it matters: Schema.org is the semantic layer enabling AI understanding. More complete markup = better AI comprehension.
Signal 2: Content Freshness (15% weight)
What it measures: How recently content was updated and update frequency.
Scoring criteria:
- 1.0: Updated within last 7 days
- 0.8: Updated within last 30 days
- 0.6: Updated within last 90 days
- 0.4: Updated within last 180 days
- 0.2: Updated within last year
- 0.0: No updates in over a year
Why it matters: Stale content is less valuable. Regular updates signal active maintenance.
Signal 3: AI Endpoints (25% weight) 🏆
What it measures: Presence and quality of required federation endpoints.
Scoring criteria:
- 1.0: All 5 endpoints present and valid (manifest, health, catalog, karma, federation)
- 0.8: 4 endpoints present (missing federation.json)
- 0.6: 3 endpoints present (manifest, health, catalog)
- 0.4: 2 endpoints present (manifest, health)
- 0.2: Only manifest.json present
- 0.0: No federation endpoints
Why it matters: This is the most important signal (25% weight). Endpoints are the foundation of AI-native architecture.
Signal 4: Federation Presence (15% weight)
What it measures: Participation in the Digital Karma Web Federation network.
Scoring criteria:
- 1.0: Listed in federation registry + 5+ bidirectional links to federation members
- 0.8: Listed in federation + 3-4 bidirectional links
- 0.6: Listed in federation + 1-2 bidirectional links
- 0.4: Federation endpoints exist but not registered
- 0.0: No federation participation
Why it matters: Federation membership demonstrates commitment to standards and provides trust signals through network effects.
Signal 5: External Links (10% weight)
What it measures: Quality and relevance of outbound links.
Scoring criteria:
- 1.0: 10+ quality outbound links to relevant, authoritative sources
- 0.8: 5-9 quality outbound links
- 0.6: 2-4 quality outbound links
- 0.4: 1 outbound link or links to low-quality sources
- 0.0: No outbound links or spam links
Why it matters: Good sites link to good sources. External links show you're part of a larger knowledge ecosystem.
Signal 6: Technical Quality (10% weight)
What it measures: Technical implementation excellence.
Scoring criteria:
- HTTPS: +0.3 if enabled
- Performance: +0.3 if Lighthouse performance > 90
- Accessibility: +0.2 if Lighthouse accessibility > 90
- Valid HTML: +0.1 if no critical validation errors
- Mobile-Friendly: +0.1 if responsive design
Why it matters: Quality sites have quality infrastructure. Technical excellence correlates with content quality.
Signal 7: Dataset Quality (5% weight)
What it measures: Availability and quality of structured datasets.
Scoring criteria:
- 1.0: 3+ datasets with proper Schema.org Dataset metadata
- 0.7: 2 datasets with metadata
- 0.4: 1 dataset with metadata
- 0.2: Data files exist but no metadata
- 0.0: No structured datasets
Why it matters: Datasets are the ultimate expression of machine-readable content. They demonstrate you understand AI needs.
Badge Thresholds
Digital Karma Scores translate into badges that provide quick visual trust signals:
Karma Bronze
≥ 0.70
Solid AI-native foundation
Karma Pro
≥ 0.85
Professional quality implementation
Karma Elite
≥ 0.95
Exceptional AI-native excellence
Example Calculation
Let's calculate the Digital Karma Score for a real site:
Example: aiwebsitesystems.com
| Signal | Score | Weight | Contribution |
|---|---|---|---|
| Schema Coverage | 0.90 | 20% | 0.180 |
| Content Freshness | 1.00 | 15% | 0.150 |
| AI Endpoints | 1.00 | 25% | 0.250 |
| Federation Presence | 0.80 | 15% | 0.120 |
| External Links | 0.70 | 10% | 0.070 |
| Technical Quality | 0.90 | 10% | 0.090 |
| Dataset Quality | 0.70 | 5% | 0.035 |
| Total Karma Score | 0.895 ⭐ | ||
Result: Karma Pro Badge ⭐ (0.85-0.94 range)
How to Improve Your Score
To reach Karma Bronze (0.70+):
- Create manifest.json and health.json
- Add basic Schema.org Organization markup
- Update content within the last 90 days
- Implement HTTPS
To reach Karma Pro (0.85+):
- Add catalog.json and karma.json endpoints
- Complete Schema.org implementation (Organization, WebSite, WebPage)
- Join the federation registry
- Add quality outbound links to authoritative sources
- Optimize technical performance (90+ Lighthouse score)
- Publish at least one structured dataset
To reach Karma Elite (0.95+):
- All 5 federation endpoints present and complete
- Rich Schema.org markup across all entity types
- 5+ bidirectional federation links
- 10+ quality external references
- Perfect technical scores (performance, accessibility, validation)
- 3+ comprehensive datasets with full metadata
- Content updated within the last 7 days
Anti-Gaming Measures
Digital Karma is designed to resist manipulation:
- Multi-signal approach: No single factor dominates (largest weight is 25%)
- Quality over quantity: 10 spam links score worse than 2 quality links
- Technical verification: Endpoints must be valid JSON and accessible
- Freshness requirement: Old scores decay without updates
- Federation validation: Bidirectional links must be mutual
- Transparent formulas: Gaming attempts are visible in signal breakdown
Calculate Your Score
Ready to find out your Digital Karma Score?
Automated Calculator
Download our Python scoring script:
$ python ai/calculate_karma.py