The Founders
AI Playbook
A practical guide for founder-led service businesses to find workflow friction, improve one system first, and use AI without losing the human touch.
The Guide
About Douglas Hunter
Douglas Hunter is the founder of Marsten AI.
After more than 30 years helping businesses and people improve operations across real estate, finance, insurance, technology, and entertainment, he developed the Marsten AI Workflow Framework to help founder-led businesses introduce AI practically, one workflow at a time, while keeping humans in control.
Start Here
Watch This First
This short video explains the complete methodology — how every piece works together. Instead of learning random AI tactics, you will understand how to build an entire AI-assisted workflow system for your business.
What You'll Learn
- How to identify workflow friction
- How to choose the first workflow to improve
- Why most AI projects fail
- How to measure whether AI is actually helping
- Where AI can assist without replacing people
System Overview
Friction Point → Workflow Map → AI-Assisted Step → Human Review → Measure & Refine
The Methodology
Find the friction. Map the workflow. Improve one system first. Keep humans reviewing. Then measure and refine.
The Foundation
The Big Idea
AI does not fix unclear workflows.
— The Marsten.AI Philosophy
It amplifies them.
Founder Reality
Most AI projects stall because the workflow is unclear, systems are fragmented, or the team is already overloaded before AI is introduced. More tools are not the answer.
How to Use This Playbook
Do not automate everything overnight. Find where friction exists first, improve one workflow, keep humans reviewing, then measure and refine before expanding.
“Most founders do not need more AI tools first. They need to know where AI actually belongs inside their business. This playbook helps them identify the first workflow worth improving.”
— Marsten.AI
The System
The Marsten.AI Workflow Framework
Mini Friction-to-Workflow Map
Identified
End-to-End
Step Built
+ Approves
+ Refined
Every successful founder-led business that uses AI well has one thing in common: they did not automate everything at once. They found one friction point, mapped the workflow clearly, added one AI-assisted step, kept a human in the loop, and measured the result. Then they repeated the process.
Find the Friction
Before choosing an AI tool, identify where your business is actually slowing down. The 7 most common friction points in founder-led service businesses are listed below. Most founders find at least 3 that apply immediately.
01 — Lead Response
Slow replies create missed opportunities
AI assist: Prepare lead summaries and first-response drafts. Measure: Response time, booked calls.
02 — Follow-Up
Important conversations become inconsistent over time
AI assist: Draft follow-up emails and next-step reminders. Measure: Reply rate, follow-up speed.
03 — Proposals
Drafting and revisions pull founders into manual work
AI assist: Create proposal outlines, recaps, and action items. Measure: Proposal turnaround time.
04 — Meetings & Action Items
Decisions happen, but follow-through gets scattered
AI assist: Organize notes, owners, deadlines, and next steps. Measure: Task completion rate.
05 — Internal Knowledge
Information lives in inboxes, documents, or someone's head
AI assist: Make FAQs, SOPs, and policies easier to search. Measure: Time saved, fewer repeated questions.
06 — Content Creation
Consistent messaging takes more time than expected
AI assist: Turn calls, notes, and FAQs into draft content. Measure: Draft time, content output.
07 — Onboarding & Admin
Repetitive setup tasks quietly consume valuable time
AI assist: Create welcome steps, checklists, and document collection. Measure: Time to onboard, handoff speed.
Map the Workflow
Once you have identified a friction point, map the workflow end-to-end before touching any AI tool. You cannot improve what you have not clearly defined. Most founders discover the real problem only when they write it out step by step.
Why This Step Matters
AI amplifies whatever workflow it is placed inside. A clear workflow produces better AI output. A messy workflow produces faster mess. Map first, automate second.
The Golden Rule
Understand the workflow before you improve it.
You are looking for the one step where AI can assist — not replace the entire process.
Worksheet — Map Your Workflow
If you're unsure how to answer every question, that's exactly what we'll work through together during your complimentary AI Review.
Your entries are stored only in this browser and are not submitted.
Choose One AI-Assisted Step
After mapping the workflow, identify the single step where AI can do the heavy lifting while a human reviews the output. Start with one. Prove it works. Then expand.
Draft First-Response Emails
AI drafts a personalised first reply based on the lead's details. Human reviews and sends.
Summarise Leads Before Calls
AI pulls key details from the lead form and creates a one-page brief before the discovery call.
Create Proposal Outlines
AI generates a structured proposal outline from call notes. Human adds context and sends.
Turn Meeting Notes into Next Steps
AI organises raw meeting notes into owners, deadlines, and action items automatically.
Organise FAQs & Internal Knowledge
AI turns scattered docs and emails into a searchable internal knowledge base.
Draft Content from Calls & Notes
AI converts call recordings and FAQs into first-draft social posts, emails, or guides.
Keep Humans in Control
Remove friction, not humans.
AI accelerates the workflow. Humans provide the judgment. Both are required.
AI Does
- Drafts first responses
- Summarises information
- Organises and structures data
- Recommends next steps
- Generates first-draft content
Humans Do
- Review and approve all output
- Make final decisions
- Add context and judgment
- Maintain client relationships
- Refine and improve the system
Measure the Result
If you cannot measure it, you cannot improve it. Every AI-assisted workflow step should have a clear metric attached before you build it. These are the six most meaningful measurements for reducing friction in founder-led service businesses.
Response Time
How fast does your team reply to new leads and enquiries?
Follow-Up Speed
How quickly does the next touchpoint happen after initial contact?
Proposal Turnaround
How long from discovery call to proposal delivered?
Task Completion
Are action items getting done, or falling through the cracks?
Time Saved
How many hours per week does the improved workflow return to the team?
Fewer Missed Opportunities
Are warm leads converting at a higher rate after the change?
“Measure the one thing that matters most for the workflow you improved. A single clear metric is more useful than five vague ones.”
— Marsten.AI
Example Workflow
Lead Response
Slow lead response is the single most common revenue leak in founder-led businesses. A lead that waits more than an hour is significantly less likely to convert. Here is how the AI-assisted workflow fixes it.
Before
A lead comes in through the website. The founder is in a meeting. The workflow breaks because the team is not sure who should respond or what to say. Hours pass. The lead goes cold. The opportunity is lost.
After
AI immediately summarises the lead's details and drafts a personalised first response. A team member reviews, approves, and sends within minutes. A follow-up reminder is automatically scheduled.
The AI-Assisted Workflow
- Lead arrives — form submission, email, or DM triggers the workflow.
- AI summarises — key details, business type, and stated need are extracted.
- AI drafts response — a personalised first reply is prepared for human review.
- Human reviews and sends — team member approves, edits if needed, and sends.
- Follow-up reminder set — AI schedules the next touchpoint automatically.
Example Workflow
Proposals
Proposal drafting is one of the highest-friction workflows for founders. It requires pulling together notes, context, and pricing — and it always seems to land at the worst possible moment. AI can do the heavy lifting while the founder adds the final judgment.
Before
After a discovery call, the founder sits down to write a proposal from scratch. It takes 2–3 hours. The client waits days. Momentum from the call is lost. Revisions add another round of manual workflow friction.
After
AI creates a structured outline from the call notes within minutes. The founder reviews the structure, adds pricing and personal context, and sends a polished proposal the same day. Turnaround drops from days to hours.
The AI-Assisted Workflow
- Discovery call ends — notes or recording are fed into the workflow.
- AI creates outline — problem, solution, scope, and next steps are structured.
- AI drafts recap — a call summary is prepared to include in the proposal.
- Human adds pricing and context — founder reviews, personalises, and approves.
- Proposal sent same day — client receives a professional, timely response.
Example Workflow
Meetings & Action Items
Decisions get made in meetings. But action items get lost in inboxes, chat threads, and memory. AI turns unstructured meeting notes into a clear, assigned, trackable workflow — without the founder having to chase anyone.
Before
The meeting ends. Everyone has a slightly different understanding of what was decided. Action items live in someone's notebook. The workflow breaks down, follow-up is inconsistent, and the same topics come up again in the next meeting.
After
AI processes the meeting notes and produces a structured summary: decisions made, action items, owners, and deadlines. It is shared with the team within minutes. Follow-through improves. Repeat discussions drop significantly.
The AI-Assisted Workflow
- Meeting ends — notes, transcript, or recording are submitted to the workflow.
- AI organises decisions — key decisions made in the meeting are extracted and listed.
- AI assigns action items — each task is linked to an owner and a deadline.
- Human reviews summary — founder or team lead confirms accuracy before sharing.
- Summary distributed — team receives a clear, shared record of what was decided.
Self-Assessment
AI Readiness Scorecard
Use this scorecard to evaluate any workflow in your business. Score each question 1–5 (1 = not at all, 5 = absolutely). Workflows with the highest total scores are your best starting points for AI assistance.
Swipe horizontally to score all three workflows.
| Question | Workflow A | Workflow B | Workflow C |
|---|---|---|---|
| Is this workflow repetitive? Does it happen the same way every time? | |||
| Does it slow the team down or create a bottleneck? | |||
| Does it get missed or fall through the cracks often? | |||
| Does it depend too much on the founder to get done? | |||
| Can a human easily review and approve the AI output? | |||
| Total Score (out of 25) |
Your scores are stored only in this browser and are not submitted.
Action Plan
Your First AI Workflow Map
Use this worksheet to plan your first AI-assisted workflow. Complete every field before building anything. A clear workflow map produces better results and prevents the most common mistakes founders make when starting with AI.
Complete Your First Workflow Map
If you're unsure how to answer every question, that's exactly what we'll work through together during your complimentary AI Review.
Your entries are stored only in this browser and are not submitted.
Ready to Take the Next Step?
Find Your First
AI Workflow.
Book your Complimentary AI Workflow Review and Marsten.AI will help you look at your current workflows, identify where friction may be slowing your team down, and map one practical AI-assisted workflow your business can improve first.
Visit marsten.ai/schedule · 15 minutes · No pitch · Zero pressure