Content Hypothesis Ledger

Content Hypothesis Ledger

Implements: commons:hypothesis-ledger-pattern — shared append-only ledger schema, status lifecycle, and scoring operations. This skill configures the pattern for the content domain (CT-### prefix, 14-day default window, content metric whitelist).

Append-only ledger of every content action. The single source of truth for what we published, optimised, or distributed, what we predicted, and what actually happened. The daily-content-brief skill writes to it on each run. Metric data comes from daily-seo-brief output and guide inventory.

Dependencies

Load commons:identity first (for product positioning and ICP context).

Storage

/var/www/html/systemprompt-web/reports/content/data/hypothesis-ledger.md

Single markdown file. One table, append-only. Newest rows at the bottom. Never rewrite history.

Schema

| id | logged | channel | action | hypothesis | metric | baseline | window_end | result | status | notes |

Fields:

• id: CT-### sequential, zero-padded to 3 digits. Separate namespace from SEO S-###, marketing H-###, and CRM C-###.
• logged: YYYY-MM-DD the action was taken.
• channel: one of optimisation, new-guide, revision, distribution, technical.
• action: one-line past-tense description. Full context goes in data/actions/CT-###.md.
• hypothesis: the "If ... then ... within ..." statement, trimmed to one line.
• metric: exact field name from the metric whitelist (validated on append).
• baseline: numeric value of metric at the moment of logging (from latest daily-seo-brief data or guide inventory).
• window_end: YYYY-MM-DD when the hypothesis matures. Default = logged + 14d (content performance takes time to materialise due to crawl/index latency and engagement buildup).
• result: numeric value of metric at window_end. Empty while in-flight.
• status: IN-FLIGHT | SEEDED | PASS | FAIL | INCONCLUSIVE | ABORTED.
• notes: free text, <=140 chars.

Channel Definitions

Action Text Store

Full context for each hypothesis lives in:

/var/www/html/systemprompt-web/reports/content/data/actions/CT-###.md

Each file contains: timestamp, skill that logged it, guide slug (if applicable), what was changed, before/after values, target URLs, measurement notes.

Operations

content-hypothesis-ledger next-id              # Allocates next CT-###
content-hypothesis-ledger log {fields...}      # Append new row + write actions/CT-###.md
content-hypothesis-ledger in-flight            # Print rows where status=IN-FLIGHT or SEEDED
content-hypothesis-ledger maturing {date}      # Print rows where window_end <= date AND status=IN-FLIGHT
content-hypothesis-ledger score {id} {result} {status} {note}   # Close out a row
content-hypothesis-ledger stats                # Pass/fail tally by channel

log operation

Steps:

1. Allocate next CT-###.
2. Pull current value of metric from latest daily-seo-brief report or guide inventory frontmatter.
3. Validate metric is in the whitelist (see below). Reject if not.
4. Compute window_end = logged + window_days (default 14).
5. Append row to ledger with status=IN-FLIGHT or status=SEEDED.
6. Write actions/CT-###.md with context.
7. Print the assigned ID.

score operation

Steps:

1. Read the row, confirm status is IN-FLIGHT.
2. Append a new row with the same id and closed-out values. Never edit existing rows.
3. If PASS, emit a diff suggestion for content-strategy-master section 4 (Winning Tactics).
4. If FAIL, emit a diff suggestion for section 5 (Dead Hypotheses).

Metric Whitelist

Only these metric names are valid (sourced from daily-seo-brief output, GSC data, and guide frontmatter):

# Guide-level metrics (from daily-seo-brief / GSC)
guide_{slug}_sessions_7d            guide_{slug}_sessions_31d
guide_{slug}_gsc_clicks_7d          guide_{slug}_gsc_impressions_7d
guide_{slug}_gsc_ctr_7d             guide_{slug}_gsc_position_7d
guide_{slug}_engagement_rate

# Aggregate metrics (from daily-seo-brief)
guides_published_total              guides_indexed
organic_sessions_7d                 gsc_clicks_7d
gsc_impressions_7d                  gsc_avg_ctr_7d

# Distribution metrics (from lead-tracker / analytics)
web_sessions_7d                     content_referral_sessions_7d

Guide-level metrics use the guide slug with hyphens (e.g., guide_claude-code-system-prompt_sessions_7d).

Default Windows

Content hypotheses default to 14-day windows because:

• New guides take 3-7 days to be indexed by Google
• Title/meta changes take 5-14 days to affect rankings
• Distribution effects take 3-7 days to materialise in referral traffic

For new guide publications, use 21-day windows (indexing + ranking stabilisation).

Integration with Daily Content Brief

daily-content-brief calls:

1. in-flight — to show active content hypotheses
2. maturing {today} — to identify hypotheses that need scoring today
3. log — to register new actions from today's brief
4. score — to close out matured hypotheses

Separation from Other Ledgers

The content ledger uses CT-### IDs. The SEO ledger uses S-###. The marketing ledger uses H-###. The CRM ledger uses C-###. They are completely independent files in different directories. Do not cross-reference IDs across ledgers.