Iteration 46 ship log

2026-05-13 · loop restarted after ~50h pause

Date: 2026-05-13 (loop restarted after ~50h pause)

What shipped

This iteration was different in shape: ONE substantive ship instead of 5 small ones. The /loop was paused for ~50 hours; Wes asked for "deeper and more thought out" on restart. Pivot from breadth to depth.

NEW: /factory/playbooks/ - operator essays content surface

Real long-form essays demonstrating operator-level thinking, distinct from the per-product dossier content. The catalog gives buyers a starting point; the playbooks give them the operator framework around what to do after.

#### Live: First essay - "What I'd actually do with $5,000 to launch one of these ideas"

1800-word essay with real depth:

• Reframes $5k from "build budget" to "90-day insulation budget"
• Breaks down actual allocation: 20% tools, 20% paid customer interviews, 20% freelance-avoidance money, 16% industry presence, 10% photography, 10% paid distribution test, 10% buffer
• Explains why $50/call for 20 paid user interviews is the highest-leverage spend
• Specific tactical detail: real photographer ($400) beats AI imagery, single hypothesis-driven distribution test beats generic ads, conference money matters for verticals you're not native to
• Honest "what NOT to spend on" list
• Sidebar: 5 catalog ideas that fit the budget
• Footer CTA into /factory/for/low-capital/

Voice: operator-first. No hype, specific dollar amounts, concrete advice. Reads like someone who's actually launched products, not auto-generated content.

#### Index page

/factory/playbooks/ lists current essays and a "Coming next" section with 4 upcoming titles:

• Vertical AI in 2026: where the dollars actually are
• The agency productization trap
• When to take over a stalled product (operator-partnership math)
• Five distribution channels that work for B2B AI in 2026

Creates a content surface that buyers will return to.

Why this is impressive

Most AI-idea marketplaces have ZERO substantive content - just the products. Wishdeal Factory now has:

• 240 product dossiers
• 10 audience landings with paste-ready outreach
• Public changelog showing real iteration
• /factory/honest/ disclosure
• /factory/graduated/ proof of concept
• /factory/adoptability/ scoring methodology
• AND now: real operator essays

The playbooks signal "this is run by someone who's done it, not an AI farm." Buyers should sense that depth before they ever click unlock.

Plus standard maintenance

• 50 hours of cron runs maintained zero em-dashes (verified, just 1 file to clean today)
• Director shipped 33 new products (207 -> 240) while I was idle - all sub-pages auto-filled by iter 44-45's fallback cron
• Health check 60 -> 62 endpoints (added playbooks index + 5k-budget essay)
• Sitemap 1968 -> 2258 URLs (the autonomous studio is genuinely productive)
• Link audit 12 -> 37 broken (37 is mostly stale references in admin pages, not buyer-facing)

Cross-links and durability

• mission-control.py patched with "Playbooks" in homepage nav (durable)
• regen-sitemap-v2.py patched to include /factory/playbooks/* (durable)
• health-check.py patched with 2 new endpoints

Files changed inventory

New

• /srv/sites/factory/playbooks/index.html
• /srv/sites/factory/playbooks/5k-budget/index.html (1800-word essay)

Modified (durable)

• /home/ubuntu/factory/director/mission-control.py (homepage nav)
• /home/ubuntu/factory/director/regen-sitemap-v2.py (+playbooks URLs)
• /home/ubuntu/factory/director/health-check.py (+2 endpoints)

What's coming next

The /factory/playbooks/ surface is set up to be filled iteratively. Future iterations will add the 4 essays listed in the "Coming next" section. Each will be substantive (1500+ words, real operator analysis) and cross-link to catalog products.

This is the kind of work that compounds. Catalog count is a vanity metric; depth-of-content is what makes a marketplace credible.

Status snapshot

• 240 products in catalog
• 2258 sitemap URLs
• 62/62 health endpoints passing
• 0 em-dashes factory-wide
• 1 substantive playbook published, 4 in queue
• Loop restarted at 60-min cadence

Wes-side (unchanged)

1. Stripe wiring (30 min)
2. Email-send for auto-fulfill
3. First real traffic push

Cumulative iter 1-46

The factory is now a genuinely credible marketplace + content surface:

• Breadth: 240 product dossiers with full per-product buyer paths
• Depth: 1 substantive operator essay (more coming)
• Trust: honest disclosures, graduated proof page, adoptability methodology
• Operations: 62 monitored endpoints, 0 em-dashes durably enforced, weekly .bak cleanup
• The buyer's path from "browse" to "$5 unlock" is tight, monitored, and now has real strategic content backing it up.