“Fire Your Agency, Keep Your Growth” is a book about how roofers can take control of their own marketing using AI tools, the Content Factory system, and Dollar-a-Day advertising. It was written by Brad Strawbridge, Dennis Yu, Ethan Van de Hey, and George Paladichuk. And it was written with AI assisting at every stage — not generating, but amplifying the real expertise of its authors.
This article documents how we used AI in the book creation process, what worked, what did not, and why the distinction between AI-generated content and AI-assisted content matters for anyone building authority in their industry.
The Raw Ingredients Came First
The single most important principle in this process: AI never created from nothing. Every piece of content in the book started with real raw material from the authors themselves. Brad Strawbridge appeared on the Coach Yu Show in March 2026, where he discussed scaling Capital City Roofing from $3 million in year one to a $10 million trajectory in year two. He appeared on the Hook Agency podcast in April 2026, where he walked through his Claude Co-Work workflows and how he built a 318-page website using AI without writing code. Dennis Yu brought two decades of digital marketing frameworks refined across Nike, Red Bull, State Farm, and hundreds of local businesses.
These podcast appearances, interviews, and hands-on operational knowledge became the source material. The Content Factory process we teach — Produce, Process, Post, Promote — is exactly how we approached the book itself. The podcasts were the Produce stage. AI processing those transcripts into structured chapters was the Process stage. Publishing to Google Docs for collaborative editing was the Post stage. And sharing the book and this meta-article is the Promote stage.
What AI Did Well
AI was excellent at structuring raw transcripts into organized chapters with consistent formatting. It generated first drafts of playbook sections, created tables comparing tools and costs, and proposed AI Action prompts that readers could copy and paste into Claude or ChatGPT. It was fast at cross-referencing information across multiple sources and flagging inconsistencies.
AI also excelled at creating the living SOP framework — linking each chapter’s playbook to a corresponding URL on blitzmetrics.com where readers could access updated processes. This kind of structural work would have taken weeks manually.
Where AI Failed Without Human Oversight
The biggest failure mode was voice. When AI drafted content for Brad’s chapters, it sounded like a marketing consultant, not like a roofer who used to be a pastor. Brad talks in short, direct sentences. He uses roofing metaphors. He tells stories about being on the roof and then coming home to spend 30 minutes on content. AI’s first drafts smoothed all of that out into generic business prose.
The fix was a deliberate tuning process: we reviewed both of Brad’s recent podcast appearances line by line and extracted his actual phrases, stories, and patterns of speech. We then rewrote the AI-generated sections to match his voice — adding the “BRAD SAYS” callouts that contain his real words sourced from real interviews with specific dates and show names. This is the E-E-A-T standard we hold ourselves to: every claim is backed by a verifiable source, every quote is real, and every author bio reflects their current, complete professional profile.
The QA Process
We developed a Book Publishing QA Checklist specifically for this project, modeled after our existing Website QA Checklist. It runs three layers: Digital Plumbing (are the links working, is the formatting consistent), Content Architecture (does it sound like the right author, are the stories real), and Authority Signals (are the E-E-A-T proof points verifiable). Each layer is audited independently before the book ships.
The Key Diagrams and Visual Assets
Throughout the book, we referenced and embedded diagrams from the BlitzMetrics teaching ecosystem that already existed: the Content Factory 4 P’s diagram (Produce, Process, Post, Promote), the Dollar a Day Framework showing the seven-step cycle from testing at $1/day to scaling winners, and the One-Minute Video methodology. These are not stock graphics. They are the actual frameworks we use with clients, now serving double duty as book illustrations that reinforce the teaching.
The Lesson for Other Authors
If you want to use AI to write an authority book, start with what Brad calls the real work: get on a podcast and talk about what you actually know. Record yourself on job sites. Let AI process those recordings into structured content. Then go back and make sure every page sounds like you said it, not like a machine summarized it. The difference between a forgettable AI book and an authority-building one is whether the raw ingredients were real. Ours were.
Dennis Yu
Dennis Yu is the CEO of Local Service Spotlight, a platform that amplifies the reputations of contractors and local service businesses using the Content Factory process. He is a former search engine engineer who has spent a billion dollars on Google and Facebook ads for Nike, Quiznos, Ashley Furniture, Red Bull, State Farm, and other brands.
Dennis has achieved 25% of his goal of creating a million digital marketing jobs by partnering with universities, professional organizations, and agencies. Through Local Service Spotlight, he teaches the Dollar a Day strategy and Content Factory training to help local service businesses enhance their existing local reputation and make the phone ring.
Dennis coaches young adult agency owners serving plumbers, AC technicians, landscapers, roofers, electricians, and believes there should be a standard in measuring local marketing efforts, much like doctors and plumbers must be certified.