5 AI Prompts That Actually Help Your Shop
The right tools for the right job.
In our first two posts AI Is A Flashlight, Not a Map and What AI Gets Wrong About Your Shop — And Why It Looks So Right, we covered where AI gets it wrong. The URL myth. The demographic data problem. The competitor comparison gap. And the confidence problem: the fact that wrong answers look identical to right ones.
That was the foundation. Now let’s talk about where AI actually helps.
Below are five prompts you can copy, paste, and use today. Each one is designed for a real task that shop owners face regularly. And each one includes what to expect back and what to watch for, so you’re not just getting output. You’re getting output you know how to evaluate.
If you’ve read The Shop Owner’s Guide to AI in Marketing, you’ll recognize a couple of these. The guide has six sample prompts with a full framework for crafting your own. This post pulls from that and adds new ones.
One rule before we start: every prompt below includes built-in constraints. That’s intentional. Telling the AI what not to do is just as important as telling it what you want. It’s the difference between getting a useful draft and getting a confident guess.
1. Evaluate a Service Page
You have service pages on your website. Are they actually good? AI can help you see them through a customer’s eyes.
But first, a quick reality check from what we talked about in our second post What AI Gets Wrong About Your Shop — And Why It Looks So Right: not every AI tool can actually read a URL you paste in. And even the ones that can, don’t always do it. So this prompt comes in two versions.
Prompt 1a: With a URL
"Review this page [URL] as if you’re a potential customer looking for [service]. Is it clear what the shop offers? Does it answer the questions I’d have before booking? What information is missing? Base your evaluation only on what’s on the page. If you’re inferring anything, say so."
After you get the response, test it. Ask: “Did you actually access the URL I provided, or are you basing this on general knowledge?”
If the tool accessed it, you’ll typically get a direct answer like “Yes, I visited the page and can see that your headline says...” with specific details from your actual site.
If it didn’t, you’ll get something more hedged: “Based on common practices for auto repair websites...” or “I wasn’t able to access the URL, but based on typical service pages...” Sometimes it won’t volunteer this at all, and the response will just sound generic. That's your signal.
If you're getting generic output, or if you’re not sure, use Prompt 1b instead.
Prompt 1b: With pasted content
"I’m pasting the text content from my [service] page below. Review it as if you’re a potential customer looking for this service. Is it clear what the shop offers? Does it answer the questions I’d have before booking? What information is missing? If you’re inferring anything, say so. [Paste your page content here]"
A note on what to paste: copy the visible text from your page. The words your customers actually read. Headlines, descriptions, calls to action, any bullet points, or service details. You don’t need to include the code, the layout, or the design. AI can’t evaluate how your page looks visually. It can evaluate whether the words on the page are clear, complete, and convincing. That’s what you’re after.
What you’ll get back:
- A customer-perspective review of clarity, completeness, and calls to action.
- Specific suggestions for what’s missing (pricing context, process explanation, trust signals).
What to watch for:
- With Prompt 1b, the AI is only seeing words. It doesn’t know if your page has strong visuals, trust badges, a click-to-call button, or a scheduling widget. It’s evaluating the copy, not the experience. That’s still valuable, but it’s not the whole picture.
- Ignore any SEO recommendations that come back unprompted. You asked for a customer perspective, not a technical audit.
2. Draft Review Responses
Responding to every review matters. But writing unique, thoughtful responses takes time, especially when you’re doing it every day.
Shameless plug: we built a tool for this. KUKUI’s AI Review Responder generates personalized responses to your Google reviews directly inside our platform. If you’re a KUKUI client, you already have access.
If you’re not, or if you want to do it yourself in ChatGPT, here’s a prompt that works:
"I own an auto repair shop called [name]. Write 3 different responses to the following Google review. Tone should be [professional/friendly/warm]. Thank the customer specifically for what they mentioned. Keep each response under 75 words. Here’s the review: [paste review]"
What you’ll get back:
- Three variations you can choose from or blend together.
- Responses that reference specific details from the review, not generic “thanks for your feedback” language.
What to watch for:
- Read every response before posting. AI occasionally adds details that weren’t in the original review. If a customer said “great service” and the AI response says “we’re glad we could help with your transmission,” that’s a fabrication.
- Adjust the tone to match your shop. AI tends to run slightly more formal than most shop owners actually talk.
3. Brainstorm a Seasonal Promotion
You know you should be running seasonal promotions. Coming up with fresh ideas every quarter is the hard part.
The prompt:
"I own an auto repair shop in [city]. My top services are [list services]. Create 5 promotion ideas for [season/month] that I could use in email marketing and social media. For each idea, include a subject line, a brief description, and a suggested call to action. My typical customer is [describe target customer]."
What you’ll get back:
- Five structured promotion concepts with subject lines and CTAs ready to adapt.
- A mix of approaches (discount-based, educational, urgency-driven, community-focused) depending on how you describe your customer.
What to watch for:
- AI doesn’t know your margins. A promotion that sounds great in theory might not make sense for your business. Run the numbers before you commit.
- The more context you provide about your customer base, the more relevant the ideas. “My typical customer” is doing a lot of work in this prompt.
4. Declined-Service Follow-Up Call Script
A customer came in for an oil change. Your tech recommended brake work. They declined. Now what?
Most CRMs already handle the email side of this. If yours sends automated follow-ups for declined services, that’s covered. But what happens when your service advisor actually picks up the phone?
That call is where the real conversion happens. And most advisors wing it.
The prompt:
"Write a brief phone call script for a service advisor following up with a customer who declined a recommended [service] at their last visit. Tone should be [helpful/conversational/professional]. The goal is to remind them why it matters without being pushy. Include a natural way to handle ‘I’m not sure I need it’ and ‘I’ll think about it.’ Keep it under 200 words."
What you’ll get back:
- A conversational script with a clear opening, a brief explanation of why the service matters, and responses to common objections.
- Language that feels like a real phone call, not a sales pitch.
What to watch for:
- AI will sometimes use urgency language that’s stronger than the situation warrants. “This is a safety issue” might be true for brakes. It’s not true for a cabin air filter. Match the tone to the actual service.
- This is a starting point for your team, not a script to read word for word. The best advisors will adapt it to their own voice. The value is giving them a structure to start from.
5. Create Google Business Profile Posts
Google Business Profile posts are one of the most underused tools in auto repair marketing. Most shops never post. The ones that do usually stop after a few weeks.
The prompt:
"Create 8 Google Business Profile posts for an auto repair shop in [city]. Mix educational tips, seasonal reminders, service highlights, and community-focused content. Each post should be under 100 words with a clear call to action. My shop specializes in [specialties]. Write in a [professional/friendly/conversational] tone."
What you’ll get back:
- Two months of GBP content (one per week) ready to customize and schedule.
- A mix of content types that keeps your profile active and varied.
What to watch for:
- AI-generated posts need your photos. A GBP post with a stock-feeling description and no image gets scrolled past. Pair each post with a real photo from your shop.
- Review every post for accuracy. AI might suggest services you don’t offer or seasonal tips that don’t apply to your climate.
The Pattern
You’ll notice something in every prompt above: they all tell the AI what it’s working with and what it shouldn’t assume.
That’s the difference between a useful result and a generic one.
Three things that make any prompt better:
- Give it real context. Your services, your location, your customer, your tone. The more it knows, the less it guesses.
- Set boundaries. “Base your answer only on what’s on the page.” “Don’t estimate rankings or traffic.” “If you’re inferring, say so.” These constraints are what keep AI output honest.
- Start fresh. New task, new conversation. Don’t carry context from a previous chat. The AI will build on what you said before, and that’s not always what you want.
For a deeper framework on writing prompts, including where to find the real data that makes them work, check out The Shop Owner’s Guide to AI in Marketing. It’s free, and Chapter 4 is built around this.
What’s Next
This post gives you prompts that produce useful output. The next one tackles what happens when AI writes the content itself.
Can you use AI to write your website pages? Your blog posts? Your emails? And what does Google actually think about AI-generated content?
Short answer: it’s not what the headlines say. We’ll break it down.
In the meantime, if you missed the first two posts in this series, start there. Post 1: AI Is A Flashlight, Not a Map and Post 2: What AI Gets Wrong About Your Shop — And Why It Looks So Right and White Paper: The Shop Owner’s Guide to AI in Marketing.
Heather Myers is the Chief Technology Officer at KUKUI, where she builds marketing and customer engagement technology for independent auto repair shops. Before joining the automotive technology space, she built information systems for public and academic libraries.
This is the second post in our ongoing series,
AI Is a Flashlight, Not a Map. New posts publish every two weeks.
Categories
Archives
