A repeatable 15-minute ChatGPT workflow for writing Twitter thought-leadership threads — with the exact prompt, the 6-tweet arc structure, and a real example thread that pulled 42k impressions in a niche B2B account. Plus the editing shortcut that turned my worst tweet into the best.

Forget 30-page brand books. The voice your readers actually hear lives in the posts you've already shipped. Here's the exact prompt, the 12 dimensions I measure, and the workflow to turn your archive into a one-page style guide that AI tools will actually follow.

A 5-step transcript-to-post pipeline that converts spoken video into written blog content while keeping the speaker's voice — and the specific prompts that do the work.

Google's PAA box shows 6-8 questions. The real PAA universe has 80-150, scattered across Reddit, Quora, and forums. Here's a 4-step workflow that mines them all in 15 minutes.

A practical 6-step workflow for producing a full Google Performance Max asset group — 20 images, 15 copy slots, logos — in under an hour, using Claude for briefs and copy and Midjourney for visuals. Includes actual prompt templates, Midjourney batching tricks, and the three things that break first.

A 6-step workflow that uses Anthropic's Computer Use to drive a rank tracking dashboard at 6 AM, diff against yesterday, and post change alerts to Slack. 90 minutes to set up, ~$0.04 per run.

The 30-second Gemini prompt that tells you the right sample size before you launch any AB test, plus the formula behind it and the failure modes it prevents.

A practical workflow for using Google's NotebookLM as a marketing research analyst — uploading competitor PDFs, G2 review dumps, and analyst reports, then querying them like a real research analyst. Includes the exact setup steps, the 6 prompts that actually return usable research, the Audio Overview trick, and the three gotchas to watch for.

Seventh Sense, Klaviyo Smart Send, and the other AI send-time tools all need thousands of engaged subscribers before they get useful. Here's what to do in the meantime — 5 practical methods that work on lists under 2,000, with the prompts and Excel formulas I use to find them.

A production-grade 4-agent content pipeline built in n8n. Each agent has a single job, a structured output, and a retry path. Includes the full workflow JSON you can paste into n8n, the prompts I use for each agent, the state schema that flows between them, and the failure modes that have actually cost me time.

A step-by-step Notion database setup for storing, versioning, and sharing marketing prompts across a team — schema, categories, version control, permissions.

Most LTV models are either too simple (AOV x N) or need a data team and a Python notebook. This is the 5-input spreadsheet model I have used across DTC, SaaS, and subscription businesses since 2019 — with the exact ChatGPT prompt that builds a custom version for your business in under two minutes.