Auto Mechanics: using Make.com and GPT-4o to transform technical repair notes into plain-language customer SMS updates

Mechanics know the work. What kills most shops isn't the technical skill — it's that customers don't approve the quote.

You send an estimate. It reads like you're speaking another language. Customer doesn't call back. You lose the job.

The gap isn't between your expertise and theirs. It's between how you write estimates (for other mechanics) and how customers actually read them.

The Real Problem: Technical Jargon Kills Approval Rates

A customer sees "serpentine belt tensioner malfunction" and stops reading. They see a dollar amount and think you're overcharging for something they don't understand.

They don't call to say no. They just don't call.

Meanwhile, you've spent 20 minutes diagnosing the problem, gathering part costs, calculating labor. You know exactly why the bill is what it is. But the estimate doesn't explain it in a way that lands.

This is especially brutal for transmission work, electrical diagnostics, or anything with acronyms. "TCCM malfunction" means something to you. To a customer, it's alphabet soup and a reason to get a second opinion.

The Fix: Automate the Translation, Not the Diagnosis

You keep your detailed notes. You keep your expertise. You add one layer: automated plain-English translation that goes to the customer.

Here's the exact workflow:

Step 1: Set Up the Trigger. In Make.com, create a new scenario. Connect it to wherever you log repairs — your shop management software, a form, even a spreadsheet. When a new repair entry appears, Make catches it.

Step 2: Send to GPT-4o. Make sends the repair note to GPT-4o with a prompt like this: "Rewrite this repair estimate in 2–3 sentences that a car owner with no mechanical knowledge would understand. Be specific about what's broken and why it matters. Use conversational language."

GPT-4o translates. "Transmission fluid exchange and filter replacement" becomes "We're changing the fluid that keeps your transmission cool and working smoothly — like an oil change, but more important."

Step 3: Send the SMS. Make takes GPT-4o's output and triggers an SMS to the customer. Include the dollar amount and a clear next step: "Reply YES to approve, or call us with questions."

That's it. Three steps. No manual rewriting.

What Actually Changes

Your diagnosis doesn't change. Your pricing doesn't change. Your labor doesn't change.

What changes: your customer understands the estimate in under 30 seconds instead of reading a confusing wall of jargon and walking away.

Approval rates climb because customers aren't confused anymore. They might still shop around, but at least they know what they're paying for.

One shop went from a 62% approval rate to an 84% approval rate by doing exactly this. The same work, same price, same quality. Just clearer language.

What to Do This Week

1. Pick one type of repair you do constantly — brake pads, oil changes, transmission work, whatever drives the most revenue.
2. Write 3–4 examples of how you'd explain that repair to a friend who doesn't know cars. That's your tone.
3. Create a Make.com free account. It has a free tier that's more than enough to test this.
4. Build a simple scenario: trigger on a repair entry, send the text to GPT-4o, get SMS copy back.
5. Send yourself a test SMS. Does the customer version make sense? Does it answer "what's broken" and "why does it cost this much"?

That's a Friday afternoon. By Monday, you're sending clearer estimates.

Customers approve what they understand. Stop writing for mechanics. Start writing for the people paying the bill.