The Problem You Already Know
You open ChatGPT. You type: "Write a response to this Google review." It spits out something generic. You fix half of it. You paste it. Tomorrow you do it again — and ChatGPT has forgotten everything.
Your pricing. Your service area. The way you talk to customers. Gone.
That's not an AI problem. That's a systems problem. You're using a tool that forgets. We're going to build one that remembers.
What You're Building
In the next 20 minutes, you'll build a Google review response system. Not a prompt. A system — one that:
- Remembers your business name, services, pricing, and service area
- Writes in your voice, not generic AI-speak
- Catches mistakes before they go live (wrong pricing, promises you don't make)
- Works every time without re-explaining anything
By the end, you'll paste in a Google review and get back a response that sounds like you wrote it — because the system knows how you write.
Step 1: Give It Memory
The biggest frustration with ChatGPT? Starting from zero every conversation.
Here's the fix. Create a file called business-profile.md with your business details:
# Business Profile
**Company:** [Your Business Name]
**Services:** [e.g., Residential HVAC — installs, repairs, maintenance]
**Service Area:** [e.g., Baltimore metro, 30-mile radius]
**Hours:** [e.g., Mon-Fri 7AM-5PM, emergency available]
## Pricing Notes
- Service call: $89 diagnostic fee (waived with repair)
- No pricing over reviews — always direct to phone
- Never promise same-day unless confirmed with dispatch
## Voice
- Professional but warm — like talking to a neighbor
- First name only (never "Dear Valued Customer")
- Mention specific work when possible
- Keep it under 4 sentences
This is memory. Instead of re-typing your business details every time, the system reads this file and already knows who you are.
Step 2: Write the Instruction
Now create the system prompt — the instruction that tells the AI what to do with every review:
You are a review response writer for {{business_name}}.
When given a Google review, write a response that:
1. Thanks the reviewer by first name
2. References the specific work mentioned
3. Keeps the tone matching the voice profile
4. Stays under 4 sentences
5. Never quotes pricing — direct to phone for estimates
6. If negative: acknowledge, don't argue, offer to make it right
Read the business profile before every response.
Notice rule #5. That's control — a constraint that prevents the AI from doing something that would hurt your business. Without it, the AI might cheerfully quote your rates in a public review.
Step 3: Test It
Paste this sample review into your system:
"Mike came out same day to fix our AC. Unit was blowing warm air and he had it fixed in about an hour. Fair price too. Would definitely recommend."
A bad response (generic AI, no system):
Thank you for your wonderful review! We strive to provide excellent service to all our customers. Your satisfaction is our top priority. We look forward to serving you again!
Nobody talks like that. Your customers can smell AI from a mile away.
A good response (your system, with memory + instruction):
Mike appreciates the shout-out, Sarah! Glad he could get your AC back to cold air quick — same-day calls like yours are exactly what we train for. Thanks for trusting us with your home. We'll be here when you need us.
The difference? The system knew your voice. It knew to reference the tech by name. It knew not to mention pricing. It didn't "strive" or "endeavor" — it talked like a real person.
What Just Happened
You built a system with three of the four components every AI system needs:
| Component | What It Does | What You Built |
|---|---|---|
| Memory | Persists your business context | business-profile.md |
| Instruction | Tells the AI what to do | System prompt with rules |
| Control | Prevents mistakes | "Never quote pricing" rule |
| Flow | Multi-step automation | (Chapter 7 of the book) |
This same pattern works for estimates, GBP posts, follow-up emails, and social content. Different instructions, same architecture.
Where This Goes Next
What you built works for one task. The book gives you two working systems — review responses and GBP posts — then teaches you the framework to design your own (Chapter 13). Estimates, follow-ups, social content — same architecture, your business rules.
Chapter 7 adds flow — where one system's output feeds another. That's where "20 minutes of setup" turns into "hours saved every week."
The contractors who figure this out first aren't just saving time. They're building a competitive advantage that compounds — because their systems get better every month while competitors are still copy-pasting from ChatGPT.
88% of successful contracting shops use AI. The difference between them and the 27% of struggling shops? Systems, not prompts.
— Jobber 2026 Industry Report