Making Money With AI

Here is the Scorecard. Give each item a 1-5 score (1 = weak fit, 5 = strong fit). Multiply where noted, then total the points.

1. Customer Pull (1-5): Score how clearly the output solves a customer problem you already hear. Example: If gym members constantly ask for meal ideas, “meal plan help” scores high. If they never mention marketing copy, that scores lower.

2. Workflow Fit (1-5): Score how similar the work is to what you already do every week. If you already write weekly class descriptions, AI can draft those quickly. If you do deep technical design work, the fit may be lower.

3. Time-to-First-Value (1-5): Score how soon you can deliver a usable result. An AI email draft you can send today scores higher than an AI system that needs weeks of data cleanup.

4. Money Impact Path (1-5): Score how directly the use case connects to money outcomes you track. For a gym, that might mean booked sessions, retention, or fewer no-shows - not “brand awareness.”

5. Risk and Compliance Cost (1-5, but invert it): Score how risky or messy the use case is. If the output could offend customers or violate rules, score it low and then subtract its effect (the goal is to avoid costly surprises).

How you calculate it: - Fit-to-Money Score = (Customer Pull + Workflow Fit + Time-to-First-Value + Money Impact Path) × Risk Modifier - Risk Modifier: Use 1.0 if Risk and Compliance Cost scores 4-5, 0.8 if it scores 3, and 0.6 if it scores 1-2.

This structure matters because it prevents two classic traps: chasing “cool AI” (high novelty, low customer pull) and chasing “perfect AI” (high ambition, slow time-to-cash).

---

Putting the Scorecard into practice with Talia’s gym

Talia, 34, owns a local gym. She runs classes, handles member questions, and posts updates to keep people engaged. She likes AI, but she also hates wasting time on drafts that nobody uses.

She looks at four AI use cases and scores them. She keeps her numbers simple: she wants something she can try within 7 days and measure within 14-30 days.

Step-by-step: score and pick the best use case

1. Write down your real customer questions (not what you wish they asked). Talia pulls ten recent messages from her phone and email: “What should I eat on rest days?”, “Can I join if I’m sore?”, “Do you have a beginner plan?”, “I missed a class - what now?” Expected outcome: you get a list of customer pull signals you can score honestly.

2. List 6-10 AI use cases you can start without rebuilding your business. Talia brainstorms quickly, then narrows: - Draft responses to member questions in her tone - Generate beginner workout plans from a short intake form - Turn her class schedule into member-friendly weekly reminders - Create simple meal plan prompts for common goals - Draft social captions from her own notes Expected outcome: you stop debating and start evaluating.

3. Score each use case using the Fit-to-Money Scorecard (1-5 each). She scores like this: - AI replies to member questions Customer Pull: 5 (people ask constantly) Workflow Fit: 5 (she answers messages now) Time-to-First-Value: 5 (she can test replies immediately) Money Impact Path: 4 (fewer drop-offs, faster responses) Risk and Compliance Cost: 4 (low risk if she keeps answers grounded) Risk Modifier: 1.0 Score: (5+5+5+4)×1.0 = 19

• AI beginner workout plans from intake form Customer Pull: 4 (beginner support matters) Workflow Fit: 4 (she designs plans now) Time-to-First-Value: 3 (needs her template rules) Money Impact Path: 5 (better onboarding increases retention) Risk and Compliance Cost: 3 (moderate risk if rules get sloppy) Risk Modifier: 0.8 Score: (4+4+3+5)×0.8 = 13.6

• AI social captions from notes Customer Pull: 3 (nice-to-have) Workflow Fit: 3 (she writes sometimes) Time-to-First-Value: 5 (easy to test) Money Impact Path: 2 (hard to connect to revenue fast) Risk and Compliance Cost: 4 Risk Modifier: 1.0 Score: (3+3+5+2)×1.0 = 13

• AI meal plan outputs Customer Pull: 4 Workflow Fit: 2 (she doesn’t build full meal plans today) Time-to-First-Value: 4 Money Impact Path: 3 Risk and Compliance Cost: 1 (high risk if nutrition advice gets too specific) Risk Modifier: 0.6 Score: (4+2+4+3)×0.6 = 8.4

4. Pick the top use case that scores highest AND you can deliver safely. Talia chooses AI replies to member questions first because it scores highest and it fits her workflow. She keeps it safe by using her existing answers as the base and only letting AI draft, not invent medical or diet advice.

5. Define one simple “proof” metric before you start. Talia sets a target: reduce unanswered messages after 6pm and speed up first response times. Expected outcome: you measure time-to-cash signals, not vibes.

Quick checklist: Talia’s “start this week” setup

• Pick your top 1-2 AI use cases using the score. - Pull 10-20 real customer questions you already received. - Create a response style rule (short, friendly, and grounded in your gym’s actual policies). - Test in a small batch (send 10 drafts, compare to your usual responses). - Track one metric for 2 weeks (example: “first response within 2 hours” or “fewer unanswered messages”).

This is how you match AI to your customers and your cash timeline. You do not guess. You score, you test, you measure.

---

What to watch for when you choose AI use cases (and how to avoid slow money)

Even good ideas can stall if you miss one key detail: AI outputs must connect to real decisions your business makes. Watch for these edge cases.

Over-scoring “easy to do” instead of “easy to sell” Do this: Score Money Impact Path based on a real outcome you already track, like booked sessions, retention, or cancellations you can reduce. Not this: Score impact based on “engagement” or “brand” when you cannot link it to revenue within a month. Fix: Force yourself to write one sentence: “This use case helps me get X by improving Y.”

Building the wrong “first version” Do this: Start with a limited output that you can review quickly, like “draft replies” or “first-pass workout outlines” using your existing templates. Not this: Start with a full automation that decides pricing, health claims, or membership eligibility. Fix: Ship a draft you control. You keep approval, and you learn what customers actually need.

Letting AI drift away from your real policies Do this: Give AI a small rules list: your gym’s membership rules, class rules, and what you never promise. Not this: Copy-paste random posts and hope AI “figures it out.” Fix: Create a “policy block” you reuse in every prompt and response. Talia keeps it to bullet points so she can update it in minutes.

---

Your chapter roadmap: score, test, and earn time-to-cash

Here is the progression to keep you moving week by week. You can follow it even if you feel new to AI.

• Day 1-2 (Score): Write 6-10 AI use cases and score each one using the Fit-to-Money Scorecard. - Day 3-4 (Prepare): Collect 10-20 real customer questions and write your response style rule. - Day 5-7 (Test): Generate drafts for a small batch, review them, and send the ones that match your tone and policies. - Week 2-4 (Measure): Track one proof metric tied to customer experience and a money-adjacent outcome.

Credibility matters, so I will say this plainly: the first time I tried AI for a business task, I picked a tool that impressed me instead of one that matched the customer questions I saw every day. The outputs looked good, but they did not change the one thing that mattered - how fast we closed the next customer conversation. That mistake taught me to score ideas by customer pull, workflow fit, and time-to-first-value, then protect the business with risk rules.

If you want AI to earn money, you need a use case that earns trust quickly and connects to a decision you already make.

Before you move to the next chapter, pick one idea you currently have in your notes and score it right now using the Fit-to-Money Scorecard. Then choose the highest-scoring one to test first. That single decision will set the pace for everything you build next.

End of chapter one. 4 more chapters in the full book.