Rule of thumb: if you cannot explain a pillar in one sentence to a stranger, it is too vague.

Ranked backlog

A single prioritized list of content ideas, scored by:

``` score = (relevance_to_pillar × 3) + (proof_available × 2) + (timeliness × 1) ```

Each item has: one-line hook, target pillar, platform (X / LinkedIn / both), proof artifact (screenshot, metric, link), and estimated effort (S / M / L).

Recurring series

Named series create audience expectation and reduce creative overhead:

• "Monday build log" — what shipped, what broke, what is next
• "Friday link drop" — 3 curated links with micro-takes
• "Deep dive Thursday" — long-form LinkedIn article or X thread

Publishing workflow

1. Capture — dump raw ideas into backlog within 2 minutes of occurrence
2. Score — weekly 15-min session to rank top 7 backlog items
3. Draft — write post body, attach proof artifact, tag pillar
4. Platform-fit — adapt tone and format (X: punchy, threaded; LinkedIn: narrative, paragraph)
5. Schedule — slot into calendar; mornings (7–9 AM local) for X, Tue/Wed/Thu for LinkedIn
6. Publish + engage — post, then spend 15 min replying in the first hour
7. Measure — weekly review of impressions, engagement rate, profile visits, follower delta

Workflow

Step 1 — Define pillars

Write 3–5 pillars. For each, list 3 example post ideas to validate the pillar is fertile.

Template:

``` Pillar: [Name] One-liner: [What this pillar is about in ≤15 words] Example posts:

1. [Idea]
2. [Idea]
3. [Idea] ```

Step 2 — Seed the backlog

Brainstorm 15–20 ideas. Score each. Sort descending. The top 7 become your first two weeks of content.

Step 3 — Set cadence

Map days of the week to pillars. Example:

Step 4 — Draft in batches

Block 90 minutes on Sunday to draft 5 posts from the top of the backlog. Store drafts in a shared doc or scheduling tool (Typefully, Buffer, or native drafts).

Step 5 — Weekly review

Every Friday, spend 15 minutes:

1. Record metrics: impressions, engagement rate, follower delta
2. Note top-performing post and hypothesize why
3. Promote 2 backlog ideas that relate to what performed
4. Archive stale ideas (>4 weeks old, no longer timely)

Examples

Example pillar set for an AI engineer

Example backlog item

``` Hook: "I replaced 400 lines of LangChain with 50 lines of raw SDK calls. Here's what I learned." Pillar: Sharp take Platform: X thread + LinkedIn adaptation Proof: Before/after code screenshots, latency benchmarks Effort: M Score: (3×3) + (2×2) + (1×1) = 14 ```

Example weekly metrics log

``` Week of Mar 24: X impressions: 42,000 (+18%) X engagement rate: 4.2% LinkedIn impressions: 8,500 (+12%) LinkedIn engagement rate: 6.1% Follower delta: +340 (X), +85 (LinkedIn) Top post: "400 lines of LangChain" thread (X) — 12K impressions Hypothesis: Contrarian takes with code screenshots outperform pure text Action: Promote 2 similar backlog items to top of queue ```

Decision tree

1. Have I defined pillars? No → Step 1. Yes → continue.
2. Is my backlog scored and sorted? No → Step 2. Yes → continue.
3. Do I have a weekly cadence mapped? No → Step 3. Yes → continue.
4. Are drafts batched ahead of time? No → Step 4. Yes → continue.
5. Am I reviewing metrics weekly? No → Step 5. Yes → maintain and iterate.
6. Is engagement plateauing? Yes → rotate one pillar, test a new series format.
7. Am I burning out? Yes → reduce to 3 posts/week, keep only highest-ROI pillar.

Edge cases and gotchas

• Pillar overlap — if two pillars produce identical posts, merge them. Four strong pillars beat five blurry ones.
• Platform tone mismatch — do not copy-paste X posts to LinkedIn verbatim. X rewards density and edge; LinkedIn rewards narrative and vulnerability.
• Vanity metrics — impressions without profile visits or follower growth mean the content entertains but does not convert. Tighten the hook-to-CTA path.
• Consistency > virality — one viral post followed by two weeks of silence loses more followers than it gains. The system is the asset.
• Proof debt — if you have no screenshots, metrics, or artifacts, you are not building in public — you are opining. Prioritize posts where you have receipts.
• Engagement window — the first 60 minutes after posting on X are critical for algorithmic reach. Do not post and disappear.
• LinkedIn algorithm shift — LinkedIn currently rewards native documents (PDFs), carousels, and long text posts over external links. Adapt format accordingly.
• Batch rot — drafts older than 10 days lose timeliness. Re-score before publishing.

Evaluation criteria