Brick 1
Framework-as-Diagnostic Overlay
Greg's 7-tactic framework overlaid on Yuri's current infra: | Tactic | Greg's framework says | We have | Verdict | |---|---|---|---| | 1. MCP server as sales team | Publish to Smithery/MCPT/OpenTools for $0 CAC discovery | Zero — neither BizBuilder/KPDD nor Solacian have an MCP server | **REAL GAP** — but applicability depends on whether the product answers a queryable question; Solacian (Maze-dissolving AI) plausibly does, KPDD (PMF discovery) plausibly does | | 2. Programmatic SEO at 10K-page scale | Next.js + Firecrawl + AI content for "best X for Y" patterns | Zero programmatic pages; KPDD/Solacian don't have SEO surface | **REAL GAP** for both; needs a keyword pattern decision before any code | | 3. Free tool as top of funnel | Grader/analyzer/calculator, instant value, share-driven loop | KPDD itself is arguably a discovery tool but is gated by signup, not free-tool-grade | **PARTIAL GAP** — KPDD could expose a free analyzer slice; Solacian could expose a free "Maze diagnostic" | | 4. AEO (Answer Engine Optimization) | Top-20 questions + FAQ schema + monitoring via Otterly/Profound | Zero structured FAQ surface on either product site | **REAL GAP** for bo…
Source: src/lib/bricks/sources/greg-isenberg-bootstrap-distribution.md
Brick 2
Strategy 2: Programmatic SEO (10,000 Pages)
The thesis: create 10,000 SEO pages in 48 hours via keyword patterns + structured data + AI-generated unique content. Math model: - 10,000 pages × 30 visits/month each = 300,000 monthly visitors - 2% conversion = 6,000 conversions/month - $10 each = **$60,000/month from pages built once** - Caveat: 30 visits/month doesn't happen overnight; SEO compounds over time Critical: content must not feel like AI. Lots of optimization needed. Start with a few pages, scale once quality is right. "Press one button" myth is rejected explicitly. Playbook (start this week): 1. Pick a keyword pattern: `[product type] for [niche]` or `[service] in [city]`. Examples: "CRM for dentists", "roofing in Miami" 2. Build dataset via Firecrawl (scrapes + clean structured data) or existing databases 3. Create page template in your framework: Next.js, Webflow, WordPress — pick what you use 4. Use AI to generate unique paragraphs per page (not variable swaps; high-quality content) 5. Human-in-the-loop editing on a sample 6. Publish 100 pages as MVP 7. Monitor indexation 8. Scale once indexed Applies to: services, SaaS, apps, boring businesses, agencies. Programmatic SEO is still under-tapped.
Source: src/lib/bricks/sources/greg-isenberg-bootstrap-distribution.md
Brick 3
B043 — Record your process → transcribe → that's the lead magnet
Record your process → transcribe → that's the lead magnet. **Note:** Manual by design — requires founder's attention. Cut-adjacent for platform.
Source: _reference/bricks/README.md
Brick 4
Flow 1: SEO Content Machine
Research keywords -> scrape page-1 -> write with perspective -> publish via CMS API -> track GSC -> optimize -> batch all keywords.
Source: src/lib/bricks/sources/gtm-engineering-flows-combined.md
Brick 5
Strategy 5: Viral Artifacts (Make Outputs Shareable)
The thesis: Spotify Wrapped gets 100M shares every December. Everyone shares it. It says something about their identity. Build the equivalent for your product. Examples in the wild: - GitHub contribution graph → devs brag about green squares on Twitter - Stripe Atlas incorporation milestone → founders tweet "just incorporated" / "5 years ago today" - Duolingo daily streak → users brag about 365-day streaks - (Snapchat streaks were arguably the original) Look at social products for design psychology, then bring it into whatever you're building. Playbook (start this week): 1. Identify the output or milestone your user would screenshot and share 2. Make it beautiful, shareable, branded — logo subtle but present (big logo = no one shares someone else's logo; it should feel about THEM) 3. Add a share button that prefills the post with the artifact 4. Every share = free impressions to your exact target audience B2B works too. B2B are people. They share in Slack or Teams. Just ask: what would they share with the group you want more of?
Source: src/lib/bricks/sources/greg-isenberg-bootstrap-distribution.md
Brick 6
1.3 Code-Generated Ad Creatives (Cody Schneider method)
Source: [[../knowledge-env/synthesized/marketing-machine-10-parallel-agents.md]] **When to use**: zero budget for design, need to test messaging variations (not visual polish), finding the winning angle/pain point before investing in production **Process**: 1. Pick a reference ad format (Facebook Ads Library for competitor research, or "before/after" template) 2. Claude Code builds the ad as a React component (1080x1080px) 3. Research pain points via Perplexity API (scrape Reddit, YouTube, Twitter for ICP language) 4. Bulk-generate text variations: titles + paragraphs mapped to pain points 5. html-to-canvas converts React components to PNG 6. Download as zip, bulk upload to Facebook Ads API as drafts 7. Run $3-5/day per variation, 3-5 days 8. Analyze CPM/CPC via data warehouse (Graph MCP or CSV export) 9. Kill high-CPM losers via Facebook Ads API 10. Winners get promoted to dedicated ad set with CPA budget **Two schools of thought on creative quality**: - School A: "Scroll-stopping creative first" - use Nana Banana Pro / Kling AI for visuals - School B: "Messaging first, visuals later" - ugly code-generated ads that speak to pain points, find the 1-3 winning …
Source: src/lib/bricks/sources/perf-marketing-playbook.md
Brick 7
1.4 Autonomous Ad Optimization Loop (Cody method)
Source: [[../knowledge-env/synthesized/gtm-engineering-flows-combined.md]] **When to use**: running 50+ ad variations, need daily kill/promote cycle without manual monitoring **Process**: 1. Test campaign runs new creative constantly (bulk-generated) 2. Daily cron job: pull performance data from data warehouse 3. Agent identifies high-CPM/low-performing ads 4. Auto-pause losers via Facebook Ads API 5. High performers get promoted to dedicated ad set with own CPA budget 6. Dashboard tracks the whole loop (clicks over time, CPC, demographics) 7. Morning brief via MCP: "How many new users went to homepage yesterday?" **Key insight**: Direct Facebook Ads API MCP has a pagination problem (you only see ~5% of your data). Need a data pipeline + data warehouse (like Graph) to get the full picture at scale. **Deployment progression**: - Start: manual in Claude Code terminal - Scale: deploy to Railway as always-on agent with cron schedule - Team: deploy UI to Vercel for team access - On-the-fly: spin up ephemeral Postgres on Railway for ad-hoc data analysis, spin down when done **Tools**: Claude Code, Facebook Ads API, Railway (deployment + ephemeral DB), Graph MCP (…
Source: src/lib/bricks/sources/perf-marketing-playbook.md