refactor(linkedin-studio): S30 magnitude-scrub (discrete-% class) — unsourced reach/engagement penalties -> SSOT

Hardening-class, NOT re-hardening: surgical SSOT-reconciliation of discrete percentage
penalties/declines stated as fact with no primary source in the SSOT
(references/algorithm-signals-reference.md). Same tool-grounded discipline as S27/S28
(read-and-show -> grep-confirm -> re-grep final). Re-grep surfaced drift + same-class siblings
beyond the plan's stored list; all surfaced and operator-approved before edit.

Scope: the discrete-% reach/engagement-penalty class only. The unsourced "Nx" reach/format
MULTIPLIER class (~50 instances across ~15 files) is a separate, larger pass -> deferred to S31
(operator: run everything, across multiple sessions).

HARDEN (20 edits, 7 files):
- linkedin-formats.md (5): :6 47-50% decline + :7 15%->31% feed-share -> directional; :176
  AI-comment 30%/55% -> ~45% less engagement (correlational, medium); :231/:279 hashtags -68%
  -> diminishing returns, no discrete figure.
- linkedin-growth-playbook (6): :158 47-50% decline (twin of formats:6) + :166 hashtags -68%
  + :224/:225 post-length 25%/32% + :435/:828 posting-frequency 25% -> directional, no
  discrete figure. (:221 1.17x multiplier folded in per operator approval; the rest of the
  multiplier class -> S31.)
- glossary.md (2): :91 engagement-bait -30-50% + :235 topic-gap -15-25% -> "correlate with
  lower reach, no discrete figure".
- engagement-coach.md (2): :195 55% + :455 -30%/-55% AI-comment -> ~45% less engagement
  (correlational), actively suppressed.
- post-feedback-monitor.md (1): :330 -25%/post -> "tends to split your own audience".
- ab-testing-framework.md (1): :66 hashtags -68% -> no discrete figure.
- poll-strategy-guide.md (2): :20 / :205 poll-overuse penalty -> declining effectiveness
  (directional).

KEPT INTACT (operator-locked / different class): engagement-pod + AI-slop suppression framing
(SSOT high-confidence); firsthour:112 (no number); poll:206 / poll:3 (qualitative); growth:567
conversion-rate; formats:268 list item.

VERIFY: discrete-% penalty/decline class re-grep across the 7 files -> NONE; leave-items intact;
bash scripts/test-runner.sh -> Passed 81 / Failed 0 / Warnings 0, exit 0; counts 29/19/26/6 +
v0.5.0 unchanged (.md prose only). Disposition: FIXED (20 edits, 7 files), one atomic commit,
local (push held).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016qgzo6rxthw7KuxHjn5vyE

references/ab-testing-framework.md

@@ -63,7 +63,7 @@ Organized by impact level. Always start with high-impact variables.
 
 | # | Variable | What to Test | Why It Matters |
 |---|----------|-------------|----------------|
-| 9 | **Hashtag count** | 0 vs. 3 vs. 5 | Diminishing returns; 5+ triggers -68% penalty. |
+| 9 | **Hashtag count** | 0 vs. 3 vs. 5 | Diminishing returns; no primary source for a discrete figure. |
 | 10 | **First comment** | With vs. without, link vs. context vs. question | First comment strategy can boost or confuse engagement. |
 | 11 | **Emoji usage** | None vs. minimal vs. heavy | Audience-dependent; professional audiences may penalize heavy use. |
 | 12 | **Line spacing** | Dense vs. airy | Readability matters on mobile but effect is subtle. |

references/glossary.md

@@ -88,7 +88,7 @@ Duration a user spends viewing content with ≥50% visible on screen. Posts with
 ## E
 
 ### Engagement Bait
-Prohibited engagement tactics ("Comment YES if...", "Tag someone who...", "Type 1 for...") that trigger -30-50% reach penalty. The algorithm actively detects and penalizes these patterns.
+Prohibited engagement tactics ("Comment YES if...", "Tag someone who...", "Type 1 for...") that correlate with lower reach — actively detected and suppressed (directional; no primary source for a discrete figure).
 
 **Used in:** `references/algorithm-signals-reference.md`
 
@@ -232,7 +232,7 @@ Three-question quality gate before publishing: (1) Does this help someone make a
 **Used in:** `references/content-angles.md`, `agents/differentiation-checker.md`, `agents/trend-spotter.md`
 
 ### Topical Consistency
-Posting about consistent topics within demonstrated expertise areas. The algorithm learns your domain expertise over 30+ days. Gaps >5 days trigger -15-25% reach penalty on return.
+Posting about consistent topics within demonstrated expertise areas. The algorithm learns your domain expertise over 30+ days. Consistency is a ranking input; gaps correlate with lower reach on return (directional; no primary source for a discrete figure).
 
 **Used in:** `references/linkedin-growth-playbook-2025-2026.md`, `references/algorithm-signals-reference.md`
 

references/linkedin-formats.md

@@ -3,8 +3,8 @@
 ## Critical Context: The Algorithm Revolution
 
 **The 2025-2026 Shift:**
-- Organic reach declined 47-50% for average users
-- Top 1% of creators: content rose from 15% to 31% of all feeds
+- Organic reach declined sharply platform-wide (directional; exact YoY varies by source — see `references/algorithm-signals-reference.md`)
+- Feed distribution concentrated further toward top creators (directional; no single sourced figure)
 - Algorithm now prioritizes topical authority over everything else
 - Dwell time became the golden metric
 - External links in the body correlate with lower reach (see `references/algorithm-signals-reference.md`)
@@ -173,7 +173,7 @@ The first hour after posting determines 70% of your post's total reach. See the
 4. **Comments <15 words** (Medium signal)
 5. **Reactions** (Lower signal - baseline engagement unit; see `references/algorithm-signals-reference.md`)
 
-**AI-generated generic comments reduce reach by 30% and engagement by 55%**
+**AI-generated generic comments correlate with ~45% less engagement (correlational, medium confidence) and are actively suppressed; no separate sourced reach figure — see `references/algorithm-signals-reference.md`**
 
 ### Dwell Time: The Golden Metric
 
@@ -228,7 +228,7 @@ Algorithm prioritizes content that keeps users on platform longer.
 
 **Current function:**
 - Metadata only, not discovery
-- Use 3-4 relevant hashtags (5+ hashtags = -68% reach)
+- Use 3-4 relevant hashtags (more shows diminishing returns; no primary source for a discrete figure)
 - Focus on keyword-driven SEO throughout content
 - Actual words in your post matter MORE than hashtags
 
@@ -276,7 +276,7 @@ Immediate engagement in first hour is critical for triggering subsequent waves.
 - Compound authority over time
 
 **Hashtag guidance (updated for 2026):**
-- Use 3-4 relevant hashtags (5+ hashtags = -68% reach)
+- Use 3-4 relevant hashtags (more shows diminishing returns; no primary source for a discrete figure)
 - Focus on keyword-driven SEO throughout post
 - Actual words matter more than hashtags
 - Don't rely on hashtags for discovery

references/linkedin-growth-playbook-2025-2026.md

@@ -155,7 +155,7 @@ The difference is between riding a wave (window active) and swimming steadily (p
 
 **What this means:**
 - Platform deliberately designs against virality
-- Organic reach declined 47-50% for most creators (2025-2026)
+- Organic reach declined sharply for most creators (2025-2026; directional, exact YoY varies by source — see `references/algorithm-signals-reference.md`)
 - Success requires understanding how LinkedIn evaluates expertise
 - Focus on engagement quality over quantity
 
@@ -163,7 +163,7 @@ The difference is between riding a wave (window active) and swimming steadily (p
 
 LinkedIn removed hashtag following, hashtag pages, and "Talks About" sections in late 2024. Hashtags are now metadata only.
 
-**Best practice:** Use 3-4 relevant hashtags (5+ triggers -68% reach penalty). Focus on keyword-driven SEO throughout content. See linkedin-formats.md for detailed hashtag strategy.
+**Best practice:** Use 3-4 relevant hashtags (more shows diminishing returns; no primary source for a discrete figure). Focus on keyword-driven SEO throughout content. See linkedin-formats.md for detailed hashtag strategy.
 
 ---
 
@@ -218,11 +218,11 @@ LinkedIn removed hashtag following, hashtag pages, and "Talks About" sections in
 ### Text-Only Posts
 
 **Performance:**
-- **1.17x average reach** for personal profiles
+- Solid average reach for personal profiles (directional; no reliable multiplier)
 - Excel at generating thoughtful comments
 - Optimal length: **1,200-1,800 characters**
-- Posts under 1,000 characters: **25% reach penalty**
-- Posts exceeding 2,500 characters: **32% underperformance**
+- Very short posts tend to underperform (directional; no primary source for a discrete figure)
+- Very long posts tend to underperform (directional; no primary source for a discrete figure)
 
 **Mobile cutoff:**
 - ~140 characters displayed before "see more"
@@ -432,7 +432,7 @@ Start professional dialogue, not sales pitch. Build relationships that convert o
 - "Secret to growth hacking on LinkedIn"
 
 **Cautions:**
-- Posting twice within 24 hours can reduce reach by 25%
+- Posting twice within 24 hours tends to split your own audience (directional; no discrete figure)
 - Consecutive posts of same format show decreased performance
 - Algorithm learns your posting rhythm
 
@@ -825,7 +825,7 @@ Avoid weekends for B2B content.
 ### Tactics That Now Hurt Performance
 
 **Excessive posting frequency:**
-Multiple times within 3 hours can reduce reach by 25%.
+Posting multiple times within 3 hours tends to split your own audience (directional; no discrete figure).
 
 **Link preview cards:**
 Significantly reduce distribution. Always remove if including external links. Consider adding links in comments or as text-only format instead.

references/poll-strategy-guide.md

@@ -17,7 +17,7 @@ LinkedIn polls generate high impressions but their effectiveness is declining in
 **Why most polls fail:**
 - Generic questions that don't teach anything
 - No follow-up content using the results
-- Overuse (more than 2 per month gets penalized)
+- Overuse (declining effectiveness; use sparingly — directional)
 - Options that are obviously "right answer" bait
 
 ## When to Use Polls (and When Not To)
@@ -202,7 +202,7 @@ What do you think — did the results match your expectation?
 |-----------|--------|
 | 1 per month | Optimal — each poll feels intentional |
 | 2 per month | Acceptable — space them 2+ weeks apart |
-| 1 per week | Too much — reach penalty, audience fatigue |
+| 1 per week | Too much — declining returns, audience fatigue |
 | Multiple per week | Algorithm suppression, looks like engagement farming |
 
 **Calendar rule:** Never post polls in consecutive weeks. Alternate with text, carousel, and story posts.