n8n · Case Study
Same Questions, Answered Without You
The problem
The support team was reading every ticket and writing every response by hand. Same questions, same answers, typed out again and again. Response time depended on who was at their desk.
Who this is for
Service businesses and small agencies where support lands on one person, and the answers already exist in internal docs or a help centre. Pays off when response time depends on who is at their desk rather than on how hard the question is.
What I built
An AI support system in n8n that answers from the business's own documentation, not from a generic model's guesses:
- Incoming emails classified by topic and intent
- A RAG knowledge base built on OpenAI embeddings, so answers come from the company's real docs
- Gemini Flash 2.0 generates the response from the retrieved context
- Auto-reply sends the answer without a human in the loop
Demonstration outcome
Demonstration disclosure: this is a portfolio build. The workflow behavior described here is not a measured client business result.
Support tickets answered accurately and automatically using the business's own knowledge base. The team stopped retyping the same answers.
Practical guides
Plain checklists for lead follow-up and booking workflows.
Buyer's Guide: Who Catches the Lead When Your Booking Workflow Fails
A checklist for coaches and consultants using GoHighLevel. Ask these questions before you pay anyone to build your lead-to-booked-call workflow.
The Inquiry Handoff Checklist: Evidence, Owner, Next Action
A plain checklist for coaches and consultants on GoHighLevel. Every inquiry that needs a person should arrive with proof, a named owner, and one next step.
A Response Time Log You Can Trust, and the Revenue Claims to Skip
A plain guide for coaches and consultants on GoHighLevel. Build a simple log, define one clock, report counts, and never claim revenue you did not recover.