≈2.2k tokens on demand. Your agent reads this file only when SKILL.md points to it.
Metrics Guide
Purpose: Channel-specific KPI definitions, Brand Health Score calculation, and platform algorithm insights.
Read when: AUDIT or MEASURE mode — user needs brand health diagnosis or performance tracking.
Brand Health Score (BHS)
Overview
A composite score (0-100) measuring overall brand presence across channels. Calculated per-channel then aggregated.
Calculation
BHS = Σ (Channel_Score × Channel_Weight) / Σ Channel_Weight
Channel Weights (Default)
| Channel |
Weight |
Rationale |
| GitHub |
25 |
Technical credibility foundation |
| LinkedIn |
20 |
Professional network and visibility |
| Tech Blog (Qiita/Zenn/own) |
20 |
SEO and thought leadership |
| X (Twitter) |
15 |
Real-time presence and networking |
| Conference/Talks |
10 |
Authority and community reputation |
| YouTube/Video |
5 |
Growing but optional |
| Newsletter |
5 |
Deep audience relationship |
Rule: Weights are adjustable based on user's goals. Career-focused → increase LinkedIn. OSS-focused → increase GitHub. Thought leadership → increase Blog/Talks.
Channel Scorecards
GitHub (0-100)
| Factor |
Weight |
Scoring Criteria |
| Profile README |
15 |
Exists with positioning, tech stack, featured work |
| Pinned Repos |
20 |
4-6 repos with clear READMEs, aligned with niche |
| Contribution Consistency |
20 |
Regular commits (weekly+), not just bursts |
| Repo Documentation |
15 |
READMEs, CONTRIBUTING, clear code comments |
| Stars/Forks (relative) |
10 |
Growth trend matters more than absolute count |
| OSS Contributions |
10 |
PRs to external projects in your niche |
| Community Engagement |
10 |
Issues, discussions, code reviews |
LinkedIn (0-100)
| Factor |
Weight |
Scoring Criteria |
| Headline Optimization |
15 |
Contains tech stack + domain keywords |
| About Section |
20 |
Positioning statement + quantified achievements |
| Experience (Impact) |
20 |
Results-first descriptions with metrics |
| Featured Content |
15 |
3-5 pinned items, updated quarterly |
| Posting Activity |
15 |
2+ posts/month with engagement |
| Endorsements/Recs |
10 |
Skills endorsed, recommendations from peers |
| Profile Completeness |
5 |
Photo, banner, contact info, certifications |
Qiita (0-100)
| Factor |
Weight |
Scoring Criteria |
| Contribution Score |
20 |
Total Contribution value (platform metric) |
| Article Quality |
25 |
LGTM rate per article, technical accuracy |
| Tag Consistency |
15 |
Articles cluster around niche-related tags |
| Stock Count |
15 |
Articles bookmarked for reference |
| Posting Frequency |
15 |
2+ articles/month for growth phase |
| Organization |
10 |
Active in org if employed (team branding) |
Zenn (0-100)
| Factor |
Weight |
Scoring Criteria |
| Article Likes |
20 |
Like count and growth trend |
| Book Publication |
20 |
Published Books (systematic content series) |
| Scrap Activity |
10 |
Active learning log via Scraps |
| Follower Growth |
15 |
Monthly follower increase |
| Content Depth |
20 |
Long-form, code-heavy, original analysis |
| Badge Revenue |
15 |
Badges received (audience willingness to pay) |
note (0-100)
| Factor |
Weight |
Scoring Criteria |
| Suki (Like) Count |
20 |
Engagement rate per article |
| Follower Growth |
15 |
Monthly increase |
| Content Variety |
15 |
Mix of free + premium content |
| Revenue |
20 |
Paid article sales, magazine subscribers |
| Cross-audience Reach |
20 |
Non-engineer engagement (comments, shares) |
| Posting Consistency |
10 |
Regular cadence maintained |
X / Twitter (0-100)
| Factor |
Weight |
Scoring Criteria |
| Bio Optimization |
10 |
Positioning + niche keywords |
| Follower Quality |
15 |
Industry-relevant followers vs bots |
| Engagement Rate |
25 |
Replies + RTs + likes / impressions |
| Thread Quality |
20 |
Well-structured technical threads |
| Posting Consistency |
15 |
Daily or near-daily activity |
| Network Effects |
15 |
Interactions with peers, mentions, spaces |
YouTube (0-100)
| Factor |
Weight |
Scoring Criteria |
| Subscriber Count |
15 |
Growth trend over absolute |
| Watch Time |
25 |
Total and per-video average |
| Upload Consistency |
20 |
Regular schedule maintained |
| Audience Retention |
20 |
Average % watched per video |
| Shorts Performance |
10 |
View count and conversion to subscribers |
| Community Tab |
10 |
Polls, posts, engagement |
Conference/Talks (0-100)
| Factor |
Weight |
Scoring Criteria |
| Talk Count (annual) |
20 |
2+ talks/year minimum |
| Venue Scale |
15 |
Mix of local → regional → national |
| Slide Publication |
20 |
Speaker Deck/SlideShare with views |
| Post-talk Content |
20 |
Blog recap, video recording available |
| CFP Acceptance Rate |
15 |
Track record of acceptances |
| Audience Feedback |
10 |
Ratings, comments, follow-up connections |
Platform Algorithm Insights
X (Twitter) — Signal Weights (2026)
| Signal |
Relative Weight |
Actionable Insight |
| Reply |
27× like value |
Reply to others' content actively |
| Conversation (multi-reply) |
150× like value |
Engage in technical discussions |
| Bookmark |
5× like value |
Create save-worthy reference content |
| External link (technical articles) |
Boosted (2026 shift) |
Links to dev.to, Substack, personal blogs now algorithmically promoted — combine with thread summary for best reach |
| Hashtags |
1-2 niche: +21% / 3+: −40% |
Minimal use; algorithm relies on semantic embeddings |
| Initial velocity |
Critical (30-60 min) |
Post when audience is active |
| Premium account |
~4× reach boost vs free |
Consider Premium for consistent developer content |
Note: X reversed its external link suppression policy in early 2026 specifically for article/blog links. Non-Premium accounts still see weaker median engagement on linked posts. Source: Sprout Social — Twitter Algorithm 2026
YouTube — Ranking Factors
| Factor |
Weight (2025-2026) |
Actionable Insight |
| Viewer satisfaction |
High |
Optimize for watch-through, not clicks |
| Long-term retention |
High |
Build returning viewers |
| Click-through rate |
Medium |
Thumbnail + title optimization |
| Watch time |
Medium |
8-15 min sweet spot for tutorials |
| Shorts views |
Growing |
700B+ daily views; use for discovery |
TikTok — Key Metrics
| Metric |
Priority |
Target |
| Watch time (% completion) |
#1 |
> 70% completion |
| First 3 seconds retention |
Critical |
Hook immediately |
| Engagement rate |
Benchmark |
> 3.15% (platform average) |
| Share rate |
High value |
Create shareable moments |
Instagram — Algorithm (2025-2026)
| Signal |
Impact |
Strategy |
| DM shares |
#1 ranking signal |
Create content people want to send |
| Saves |
High |
Carousel tutorials get saved |
| Reels reach |
Primary discovery |
2-4 Reels per week |
| SEO keywords |
Growing (Google index) |
Keywords in captions + screen text |
KPI Tracking Template
Monthly Report Format
## Brand Health Report — [Month YYYY]
### Overall BHS: [Score]/100 (MoM change: +/-[N])
### Channel Summary
| Channel | Score | Trend | Key Metric | Action |
|---------|-------|-------|------------|--------|
| GitHub | [N] | [↑↓→] | [key metric] | [next action] |
| LinkedIn | [N] | [↑↓→] | [key metric] | [next action] |
| Blog | [N] | [↑↓→] | [key metric] | [next action] |
| X | [N] | [↑↓→] | [key metric] | [next action] |
| Talks | [N] | [↑↓→] | [key metric] | [next action] |
### Top Wins
- [Win 1]
- [Win 2]
### Gaps Identified
- [Gap 1 + remediation]
- [Gap 2 + remediation]
### Next Month Focus
- [Priority 1]
- [Priority 2]
Vanity Metrics vs Impact Metrics
| Vanity Metric |
Why It Misleads |
Impact Alternative |
| GitHub stars |
Can be gamed, no usage indicator |
Downloads, dependent repos, contributor count |
| Twitter followers |
Bots, inactive accounts |
Engagement rate, reply quality, DM conversations |
| Page views |
Empty traffic |
Time on page, scroll depth, return visits |
| LinkedIn connections |
Mass connecting |
Profile views from target audience, inbound messages |
| YouTube subscribers |
Inactive subscribers |
Average view %, returning viewers, comments |
Rule: Always pair a vanity metric with an impact metric. AP-2 (Vanity Metrics) applies when only vanity metrics are cited.
