22d (edited) • General discussion
🎯 Case Study: How I sold a $5000 AI Automation to a catering service
Companies don’t usually come to you asking for AI.
They come to you because they’re stressed. Because they’re drowning in tedious work. Because customers want answers immediately and their inbox keeps overflowing.
What they really want is less chaos, fewer mistakes, and more time to actually run their business instead of wrestling with emails. They want someone who understands what their day looks like and can translate that into a simple, reliable system that works in the background.
That’s the gap I repeatedly step into and this month it led to a $5,000 automation project with a monthly $600 retainer on top.
It all started with a simple message. I reached out to a local catering business at an in-person event and asked if there was anything in their work that felt inefficient or repetitive, and whether we could explore fixing it together.
No pitch deck. No “AI this, AI that.” Just curiosity. They said yes immediately, and we set up a 30-minute call for the next morning. On that call, I didn’t talk about automation at all. Instead, I asked them about their day.
What slows you down? What annoys you? What do customers complain about? What do you complain about?
You can see a business owner’s face change when they start talking about the tasks they secretly hate. Their biggest frustration seemed obvious at first: “We take too long to send quotes.” But that’s never the real problem, so I dug deeper.
I used the 5 Whys framework, asking “why” again and again until we reached the real root cause. By the final why, the picture was clear: slow quotes weren’t the problem. The real issue was lost revenue and time.
They were losing leads simply because they couldn’t respond fast enough, and customers were booking competitors before they even hit “send.” That’s the kind of insight that turns a small technical fix into a real business solution.
With that understanding, I started mapping how their entire process actually works. Asking questions like: What information comes in? Where does it go? Who touches it? When does something break?
We traced what happens next: extracting details manually, chasing missing information, copying stuff into spreadsheets, calculating pricing by hand, formatting quotes, sending follow-ups. When we laid it all out, the system looked like a patchwork of manual steps stitched together by stress.
Once everything was clear, I translated it into a simple process model — something visual, high-level, and easy to understand. That became the foundation for a proper requirements document. Not a technical piece of jargon, but a simple blueprint: here’s what needs to happen, here’s how information should flow, here’s what the final output should look like.
When I sent it over, I included one line: “If this looks good, I can deliver the entire solution for $5,000.” They signed immediately, no convincing needed.
Building it in n8n was the easy part. I connected their inbox, WhatsApp, and website forms, set up an AI extraction step, wired in their pricing logic, generated beautiful auto-formatted quotes, and made everything send itself for sign-off.
But before pushing anything live, I spent a lot of time breaking it on purpose. What happens if someone leaves out the event date? What if the budget is missing? What if the email is formatted strangely? What if two leads come in at the same time?
Testing isn’t glamorous, but it’s the difference between a tool that looks cool and a system people can depend on.
When we did the demo, the owner’s reaction said everything. I showed how a customer inquiry instantly turned into a structured data record, a calculated quote, and a formatted PDF sent straight to the client all within under a minute. Their eyes widened. “This feels like we hired a new employee,” they said. We made a few small adjustments: tweak the tone of the quote email, adjust some pricing categories, fix a logo position - and that was it.
Deployment was just plugging in real API keys, connecting their actual Gmail and WhatsApp accounts, and switching the whole system from test mode to production. After everything was stable, I handed over documentation, recorded Loom videos, and created a simple guide on how to update pricing without touching the automation itself.
Then came the easiest part of the entire deal: offering ongoing support. I simply said, “I can maintain everything, ensure uptime, and add improvements whenever needed for $500 a month.” When your system becomes part of how the business runs, the retainer becomes the obvious choice.
0
0 comments
🎯 Case Study: How I sold a $5000 AI Automation to a catering service
powered by
skool.com/ai-automations-by-jan-7120
🧠Join the free community. Beginner-friendly AI learning & Premium n8n Templates. Learn how to actually make money with AI.
Suggested communities
Powered by