Published on

Notes on Marketing

Authors

Reddit Marketing Playbook

Case Study: 11M Impressions with AI-Assisted Content

From @thepatwalls (Founder of Starter Story):

"Subreddits are still such an underrated place to get early users and high signal feedback."

Results achieved:

  • 11M impressions across multiple subreddits
  • Hit front page multiple times
  • AI helped with heavy lifting of writing (real content, not spam)

Key principle: Reddit users can smell marketing. But they upvote actually helpful content.

SaaS + Reddit Case Study: 1.4M Views

From @chhddavid's Shipper SaaS:

  • 1,413,206 Reddit views
  • 18,557 search clicks
  • Launched under 6 months ago
  • Key: Build in public + genuine engagement

Reddit Lead Generation Tool

From @indexsy:

"Reddit is full of people BEGGING for solutions."

Tool that makes Reddit searchable for leads:

  • Describe your product
  • Get list of users ready to buy
  • Find people actively asking for your solution

TikTok Marketing Mastery

The AI Content Factory Stack

From @maverickecom:

Stack: Nano Banana + Fastmoss + Manus + Veo3

What it does:

  • Repurposes content automatically
  • Localizes for different markets
  • Launches across creator-style accounts
  • No actors, no products in hand, no ghost creators

Results:

  • CPMs as low as $0.10
  • No reliance on paid ads
  • 24/7 viral TikTok Shop sales

"It's so effective it feels like running Facebook ads in 2008."

TikTok Algorithm Hack: Static Slideshows

From @gaucheai:

"TikTok is broken right now. Everyone is sweating over 4K video production while anon accounts are hitting 27M views with static AI slideshows."

Why it works: Swipe Friction

  • The algorithm loves interaction
  • Every swipe counts as engagement
  • Static slideshows encourage swiping
  • No expensive equipment needed

Making $4,078/day with AI Slideshows

From @0adspend:

  • All organic, 0 ads
  • Static AI slideshows
  • "Easiest way to make money in 2026"

Nano Banana Wrappers - 50M Views/Month

From @ErnestoSOFTWARE:

  • TikTok slideshow format
  • AI home decor app concept
  • "Absolute money glitch right now"

AI Influencer Creation System

From @alexcooldev:

3-Step System:

  1. Create avatar with Nano Banana
  2. Generate script with ChatGPT
  3. Turn script + avatar into UGC with Arcads

Result: Organic growth on TikTok, YouTube, IG instead of paying $1,000 per video.

6 AI Influencers for App Marketing

From @Philo01:

  • Submitted 6 apps to iOS App Store
  • Generated 6 AI influencers to promote
  • "Pretty crazy how real these look and sound"

AI Videos That Don't Sound Like AI

From @FynCas:

The Problem: Sora 2, Veo 3.1, Kling 2.6 all have that robotic AI accent.

When swapping in ElevenLabs:

  • Sounds like a $5,000 mic
  • Sounds like a soundproof studio
  • Too polished = obviously fake

Solution: Match voice quality to content style, use natural-sounding techniques.

Copy Any Ad with AI

From @EHuanglu:

  • Upload any existing ad + your product photo
  • AI generates product ad with realistic character
  • Character holds and introduces your product
  • Video and audio are so real no one can tell it's AI

X/Twitter Marketing Strategy

Organic UGC Program - 100M+ Views/Month at $0.10 CPM

From @kobyjconrad (Sunflower App - 10,000 paying subs):

"Our cost is about $0.10 per 1,000 views. We're not out here paying a $1 CPM. Influencers are humans, just like everyone else on your team."

Results:

  • 100M+ views per month through organic UGC
  • One of the largest UGC programs
  • $0.10 CPM (vs typical $1+ CPM)

How they did it:

  • Treat influencers like teammates, not transactions
  • Get them bought in on the mission/purpose
  • Invite them to standups
  • Give them equity (cash for posts is worst way)
  • Don't need 30 people - hunt the biggest person in your industry
  • Convince them to join your team

Still using Facebook:

  • Despite being "cringe," still the largest social platform
  • 1M sub page = 100M+ monthly organic reach
  • Has direct, trackable link clicks

The Shift to Media-First Software

"5 years from now, apps are going to become a content marketing strategy. A piece of value you put out into the world in order to sell your REAL product."

Implication for marketers:

  • Apps become lead magnets, not the product
  • Monetization through attention (ads) or physical goods
  • Build massive free audience first (100M+ users)
  • Layer real business underneath the attention

$200K in 56 Weeks - Complete Process

From @AlexGroberman_:

Results:

  • $200,000 direct revenue
  • $40,000/month in SEO Stuff sales

Posting Schedule:

  • 2-3 posts per day
  • 7 days per week
  • Morning: Proof or strong point
  • Afternoon: Story

Content Ecosystem Strategy

From @neilpatel:

"Stop Blasting Everywhere. Start Building a Content Ecosystem."

The play:

  • Slice one pillar piece into native, platform-specific content
  • Dominate YouTube, TikTok, LinkedIn, and AI search
  • AI is increasingly citing video content

SEO & AI Search

Programmatic SEO - 100+ Pages in 30 Days

From @Nick_zv_ on driving thousands of leads/month with programmatic SEO:

"One page = one keyword. 100 pages = 100 keywords. 500 pages = 500 chances to rank. This isn't about gaming the system. It's about strategically covering your entire keyword universe at scale."

Winning page patterns:

  • [Service] in [City] - Local landing pages (200 cities = 200 pages)
  • [Tool] vs [Tool] - Comparison pages for high-intent searches
  • [Job title] salary in [Location] - Data pages with massive search volume
  • Best [product] for [use case] - Buyer-intent pages that convert
  • Zodiac compatibility pages (78 combinations for 2 signs, 256 for MBTI)

Pattern requirements:

  • Search volume across variations
  • Clear intent (commercial or informational)
  • Obvious data source (cities, products, job titles)

The Programmatic SEO Workflow

Step 1: Build Variation Matrix

Create spreadsheet with 4 columns:

  1. Primary variable (city names, product names)
  2. Search volume per variation
  3. Competition level
  4. Priority score

Target: 100-500 variations minimum

Data sources: Semrush, Google Keyword Planner, Screaming Frog

Step 2: Create Perfect Template

Write ONE page manually with:

  • H1 with primary keyword slot
  • 3-5 H2 sections that work for ANY variation
  • Dynamic content blocks (stats, local info)
  • Internal linking structure
  • Clear CTA placement

Example template:

H1: Best [Service] in [City]
H2: Why [City] Residents Need [Service]
H2: Top [Service] Options in [City]
H2: How to Choose a [Service] Provider
H2: [Service] Costs in [City]
CTA: Get a Free Quote

Step 3: AI Generation with Claude

Prompt framework:

I have a landing page template for [service] pages. Here's my template: [paste template]
Here's my list of variations: [paste 20 cities]

For each variation:
- Keep structure identical
- Make content unique and locally relevant
- Include 2-3 specific details about [City]
- Maintain consistent word count (800-1,200 words)
- Don't repeat same phrases across pages

Critical tips:

  • Generate in batches of 10-20 (quality drops with larger batches)
  • Always include local data points (population, landmarks, industries)
  • Manually review 10% before bulk publishing

Step 4: Add Real Uniqueness

Google hates thin duplicate content. What works:

  • Local statistics (Census data, BLS numbers)
  • City-specific FAQs
  • Nearby location links
  • Local testimonials or case studies
  • Dynamic pricing data

Good: "Austin has 1.2M residents and 47 personal injury firms" Bad: Generic copy with just city name changed

Step 5: Technical Foundation

URL structure: /personal-injury-lawyer/austin/ NOT /pages/id=48271/

Page speed requirements:

  • Same template = same assets = fast loading
  • Compress images once, reuse everywhere
  • Lazy load below-fold content

Crawl budget management:

  • Submit XML sitemap with all new pages
  • Use internal linking to distribute authority
  • Monitor indexing in Google Search Console
  • Set up proper canonical tags

Step 6: Gradual Rollout

DO NOT publish 500 pages on day one (Google will flag it):

  • Week 1: Publish 20-30 pages
  • Weeks 2-4: Add 50-100 more
  • Month 2+: Scale to full inventory

Track in GSC:

  • Indexing rate
  • Impressions per page
  • Click-through rates
  • Position changes
  • Kill pages that don't perform after 90 days

Common mistakes to avoid:

  • Zero unique content (just swapping city names)
  • No internal linking between pages
  • Publishing everything overnight
  • Ignoring page speed
  • Forgetting mobile optimization
  • No conversion tracking

Timeline reality:

  • 100 pages published in 30 days
  • 60-70% indexed within 60 days
  • First rankings appear month 2-3
  • Traffic compounds month 4+

The Most Underrated SEO Workflow

From @JulianGoldieSEO:

3-Step Process:

  1. Start in Perplexity - Complete SEO breakdown in 30 seconds
  2. Send to Notebook LM - Clusters, outlines, angles, expert insights
  3. Push to Gemini - Full blog, titles, meta description, TikTok scripts

SaaS SEO Results

From @kalashbuilds (24 hours after launch):

  • 3,000+ visitors
  • 550+ signups
  • 8 paying users

Next features:

  • Schema
  • Search engine results page
  • Web vitals
  • pSEO
  • AEO, GEO, AI search optimization

Automation & Workflows

1,200+ N8N Workflows Leaked

From @JulianGoldieSEO:

Scraped from official docs, GitHub, and hidden forums:

  • AI-powered lead generation
  • Auto content repurposing
  • Sales & CRM automation

Social Media Automation (n8n + AI Agents)

From @SimslearnAi:

Automate Instagram, TikTok, X, FB & LinkedIn:

  • Auto content generation
  • Smart scheduling
  • Analytics tracking
  • Cross-posting

Google Colab in VS Code

From @JulianGoldieSEO:

Google now lets you run Colab notebooks in VS Code, Cursor, Windsurf:

  • Free GPUs
  • Full IDE features
  • No browser restrictions

Product Video Automation

AI System for E-commerce Videos

From @recap_david:

What it does:

  • Firecrawl scrapes product images from any e-commerce site
  • AI generates product videos automatically

Results:

  • Saves ~$30K per collection shoot
  • Boosts on-site conversion rates by ~20%

AI Tools for Marketing

Content Creation Stack

  • Nano Banana - AI avatar/image generation
  • Arcads - Turn scripts into UGC
  • Veo3 - AI video generation
  • Fastmoss - Content automation
  • Manus - Content workflow
  • ElevenLabs - AI voices

Product Sourcing

From @gregisenberg:

Alibaba's Accio AI Agent:

  • Spots trends automatically
  • Turns trends into product ideas
  • Designs the products
  • Matches with real suppliers

Lead Generation

  • Reddit search tool - Find people asking for your solution
  • N8N workflows - Automated lead capture
  • TrustMRR - Research competitor channels

Key Principles

  1. Platform-native content - Don't cross-post, adapt
  2. AI assists, doesn't replace - Keep content authentic
  3. Swipe friction > production value - Engagement beats polish
  4. Consistency compounds - Daily posting beats occasional viral hits
  5. Distribution first - Build audience before product
  6. Low CPMs exist - TikTok Shop can deliver $0.10 CPMs

References

Discuss on Twitter • View on GitHub

Related Posts