AI Chat That Actually Books Jobs
Most AI chat widgets on trade and franchise websites are glorified FAQ bots that frustrate buyers and convert at under 1%. A well-briefed AI assistant does the opposite — it qualifies, books and routes ready-to-buy visitors 24/7, pulling 15–30% of website traffic into the pipeline. The difference is not the model, the budget or the agency that built it. The difference is the brief. This guide gives you the exact three-job brief, what to feed the AI, and the four numbers to measure it on.
The three-job brief
Your AI chat has three jobs, in this order: (1) answer the buyer's question honestly, (2) qualify the job by type, urgency and location, (3) book the next step — callback, estimate or visit.
Most chat bots skip job number one and jump straight to 'enter your email to chat with a human'. That is why they convert at 1% instead of 20%. Buyers do not want to fill in a form to ask a question; they want the question answered. An AI that answers honestly earns the right to ask for the contact details a moment later.
Job two is the qualification step. Four questions are usually enough: what is the job, how urgent, what suburb, are you the decision-maker. Asked conversationally, these four questions feel like a helpful chat. Asked as a form, they feel like an interrogation.
Job three is the booking step. Never end an AI conversation with 'thanks, someone will be in touch'. End it with a specific next step — a calendar slot for a callback, a sent SMS confirming the visit window, or a transfer to a live tech for an emergency.
What to feed the AI
Your Big 5 articles — cost, problems, comparisons, reviews and best-in-class — give the AI real answers to real questions. (See the Big 5 Content Plan.)
Your service area list with suburb-level detail. 'We service all of Sydney' is too vague; 'We service Inner West, Eastern Suburbs, Lower North Shore and same-day to Parramatta' lets the AI answer 'do you come to Strathfield?' correctly.
Your pricing ranges and what moves the price up or down. Without this, the AI either refuses to discuss price (and loses the lead) or invents a number (and creates a complaint).
Your booking calendar or callback workflow, so the AI can offer a real next step.
A clear 'never do this' list: never quote a fixed price, never promise same-day without checking the calendar, never disparage competitors, never claim qualifications you do not hold.
Tone, length and personality
The AI's tone should match the brand and the trade — friendly and practical for plumbers, calm and reassuring for funeral services, energetic and outcome-focused for franchise sales. Generic 'helpful assistant' tone signals automation and kills trust within two replies.
Keep replies short — 2–4 sentences maximum, with a single clear question at the end. Long replies feel like a wall of text and double the drop-off rate. If the answer genuinely needs more detail, link to the relevant Big 5 article and offer to summarise.
Handoff: when AI gives up to a human
Define explicit handoff triggers: any mention of an emergency, any complaint, any request to speak to a human, any conversation longer than 8 turns without resolution. A good AI knows when to step aside; a bad one keeps trying to handle a job it cannot. Route handoffs to a real human within 60 seconds during business hours, or to the missed-call text-back flow outside hours.
How to measure it
Track four numbers weekly: conversations started, questions answered satisfactorily (rated by the visitor or by sampled review), qualified leads captured, bookings made. If conversation-to-lead is under 15%, the brief is wrong — usually the AI is jumping to qualification before answering. If lead-to-booking is under 40%, the qualification questions are wrong or the next step is too vague. Tune the brief weekly for the first month, then monthly.
Real-world examples
Plumbing — multi-suburb Melbourne
Replaced a generic 'leave your email' chat with a three-job AI brief fed by 11 Big 5 articles. Conversation-to-lead went from 1.4% to 22%. Lead-to-booking from 31% to 52%. Net: 14x more bookings per chat conversation.
Real-estate franchise — 60 offices
Centralised the AI brief at the network level, personalised the suburb and agent lists at the office level. After-hours leads (5pm–8am) grew from 4% of total inbound to 27%, with no extra staffing cost.
Key takeaways
- Brief the AI on 3 jobs: answer, qualify, book — in that order
- Feed it your Big 5 content, not generic FAQs
- Set explicit 'never do this' guardrails
- Keep replies short, in-brand, ending with one clear question
- Measure 4 numbers weekly and iterate the brief
Frequently asked questions
Does an AI chat actually book jobs for tradies?
Yes, when briefed correctly. A three-job AI (answer, qualify, book) fed with real pricing, service area and Big 5 content typically converts 15–30% of website visitors into qualified leads, with 40–55% of those becoming booked jobs.
What is the difference between a chatbot and an AI assistant?
A chatbot follows a scripted decision tree and breaks the moment a buyer asks something it was not programmed for. An AI assistant uses a language model briefed on your content, so it answers genuinely new questions in your voice. Conversion difference is typically 10–20x.
What should I feed an AI chat for my service business?
Big 5 articles, suburb-level service area, pricing ranges with variables, booking calendar logic, and an explicit 'never do this' list. Without these inputs the AI either refuses to answer or invents answers, both of which kill conversion.
How do I measure if my AI chat is working?
Four weekly numbers: conversations started, questions answered satisfactorily, qualified leads captured, bookings made. A healthy setup runs at 15%+ conversation-to-lead and 40%+ lead-to-booking.
Keep reading
Hand-picked guides that pair with this one.
Want this installed for your business?
We'll set up the conversion tools described in this playbook — call tracking, 60-second estimator, AI chat and smart callback.
