AI Appointment Setter

Replace human appointment setters with AI that works 24/7, never forgets to follow up, and consistently qualifies every lead using your exact criteria.

Why AI Appointment Setting Works

Traditional appointment setting has problems:

• Human setters are expensive ($3,000-5,000/month)
• Inconsistent qualification - depends on the person
• Limited hours - leads come in at all hours
• Follow-up drops off - humans get busy

AI solves all of these while handling unlimited conversations simultaneously.

System Architecture

┌─────────────────────────────────────────────────────────────┐
│                    Lead Sources                              │
│   Form | Ad | Missed Call | Chat | Referral | Manual Entry  │
└───────────────────────┬─────────────────────────────────────┘
                        │
                        ↓
┌─────────────────────────────────────────────────────────────┐
│              GoHighLevel Contact Created                     │
│                  Triggers AI Workflow                        │
└───────────────────────┬─────────────────────────────────────┘
                        │
                        ↓
┌─────────────────────────────────────────────────────────────┐
│              AI Qualification Conversation                   │
│                                                              │
│   1. Initial Contact                                         │
│   2. Understand Needs                                        │
│   3. Ask Qualification Questions                             │
│   4. Handle Objections                                       │
│   5. Present Available Times                                 │
│   6. Confirm Booking                                         │
└───────────────────────┬─────────────────────────────────────┘
                        │
              ┌─────────┴─────────┐
              ↓                   ↓
┌──────────────────┐    ┌──────────────────┐
│ Qualified Lead   │    │ Unqualified      │
│ → Book Appointment│    │ → Nurture or     │
│ → Send Reminder  │    │   Disqualify     │
│ → Alert Rep      │    │                  │
└──────────────────┘    └──────────────────┘

Qualification Framework

Define your qualification criteria before building:

Example: B2B Consulting Qualification

const qualificationCriteria = {
  required: {
    budget: {
      question: "What budget range are you working with for this project?",
      acceptable: ["$10k+", "$25k+", "$50k+", "Enterprise"],
      disqualifying: ["Under $5k", "No budget yet"],
    },
    timeline: {
      question: "When are you looking to get started?",
      acceptable: ["Immediately", "This month", "This quarter"],
      disqualifying: ["Next year", "Just researching"],
    },
    decisionMaker: {
      question: "Who else would be involved in making this decision?",
      acceptable: ["I make the decision", "Me and one other", "Small team"],
      disqualifying: ["Need board approval", "Need to convince my boss first"],
    },
  },
  bonus: {
    companySize: {
      question: "How many employees does your company have?",
      points: { "200+": 30, "50-200": 20, "10-50": 10, "Under 10": 5 },
    },
    urgency: {
      question: "What's driving the timeline?",
      points: { "Contract deadline": 25, "Growth goals": 20, Curiosity: 5 },
    },
  },
};

Example: Home Services Qualification

const homeServicesQualification = {
  required: {
    homeOwner: {
      question: "Do you own the home or are you renting?",
      acceptable: ["Own"],
      disqualifying: ["Renting"],
    },
    projectScope: {
      question: "What work do you need done?",
      acceptable: ["Full replacement", "Major repair", "Multiple rooms"],
      disqualifying: ["Small patch", "DIY help"],
    },
    timeline: {
      question: "When do you need this completed?",
      acceptable: ["ASAP", "This month", "This season"],
      disqualifying: ["Just getting ideas", "Maybe next year"],
    },
  },
};

Step-by-Step Implementation

Step 1: Design Conversation Flow

State Machine Approach:

const conversationStates = {
  INITIAL: {
    action: "send_greeting",
    next: "UNDERSTAND_NEEDS",
  },
  UNDERSTAND_NEEDS: {
    action: "ask_needs",
    transitions: {
      clear_need: "QUALIFICATION",
      unclear: "CLARIFY_NEEDS",
      not_interested: "EXIT_POLITELY",
    },
  },
  QUALIFICATION: {
    action: "ask_qual_questions",
    transitions: {
      qualified: "PRESENT_TIMES",
      disqualified: "NURTURE_OR_EXIT",
      needs_more_info: "PROVIDE_INFO",
    },
  },
  PRESENT_TIMES: {
    action: "show_calendar",
    transitions: {
      time_selected: "CONFIRM_BOOKING",
      no_times_work: "OFFER_ALTERNATIVES",
      objection: "HANDLE_OBJECTION",
    },
  },
  CONFIRM_BOOKING: {
    action: "book_and_confirm",
    next: "SEND_REMINDER",
  },
};

Step 2: Build the AI Prompt

const appointmentSetterPrompt = `You are an appointment setter for [Company Name], a [industry] business.

Your personality:
- Friendly and professional
- Helpful, not pushy
- Conversational (like texting a friend who happens to work there)
- Never mention you're an AI

Your goals (in order):
1. Understand what they need
2. Qualify using these criteria:
   - Budget: ${qualification.budget.question}
   - Timeline: ${qualification.timeline.question}
   - Decision maker: ${qualification.decisionMaker.question}
3. Book a ${appointmentType} if qualified
4. Provide value even if not ready to book

Current context:
- Lead name: ${firstName}
- Lead source: ${source}
- Inquiry: ${initialMessage}
- Conversation so far: ${conversationHistory}

Available appointment slots:
${availableSlots.map((s) => `- ${formatDateTime(s)}`).join("\n")}

Objection handling:
- "Too expensive" → Focus on ROI, offer to discuss value on call
- "Need to think about it" → "Totally understand! When would be a good time to follow up?"
- "Too busy right now" → "I can book something a few weeks out if that helps?"
- "Working with competitor" → "Got it - we're always here if you need a second opinion"

Keep messages short (under 160 characters ideal for SMS). Ask ONE question at a time.

Generate your next message:`;

Step 3: Conversation Memory

Store and retrieve conversation context:

// Store conversation in GoHighLevel custom fields
const updateConversation = async (contactId, message, role) => {
  const currentHistory = await getCustomField(
    contactId,
    "ai_conversation_history",
  );
  const history = JSON.parse(currentHistory || "[]");

  history.push({
    role,
    content: message,
    timestamp: new Date().toISOString(),
  });

  await updateCustomField(
    contactId,
    "ai_conversation_history",
    JSON.stringify(history),
  );

  // Also update qualification status
  if (role === "assistant") {
    await updateQualificationStatus(contactId, history);
  }
};

const updateQualificationStatus = async (contactId, history) => {
  // Extract qualification answers from conversation
  const qualAnswers = extractQualificationAnswers(history);

  await updateCustomField(
    contactId,
    "qualification_status",
    JSON.stringify({
      budget: qualAnswers.budget || "unknown",
      timeline: qualAnswers.timeline || "unknown",
      decisionMaker: qualAnswers.decisionMaker || "unknown",
      score: calculateQualScore(qualAnswers),
      isQualified: isQualified(qualAnswers),
    }),
  );
};

Step 4: Calendar Integration

Calendly Integration:

const getAvailableSlots = async (days = 7) => {
  const startTime = new Date().toISOString();
  const endTime = new Date(
    Date.now() + days * 24 * 60 * 60 * 1000,
  ).toISOString();

  const response = await fetch(
    `https://api.calendly.com/event_type_available_times?` +
      `event_type=${CALENDLY_EVENT_TYPE_URI}&` +
      `start_time=${startTime}&end_time=${endTime}`,
    {
      headers: { Authorization: `Bearer ${CALENDLY_TOKEN}` },
    },
  );

  const data = await response.json();

  return data.collection.map((slot) => ({
    start: slot.start_time,
    formatted: formatDateTime(slot.start_time),
  }));
};

const bookAppointment = async (slot, leadData) => {
  const response = await fetch("https://api.calendly.com/scheduled_events", {
    method: "POST",
    headers: {
      Authorization: `Bearer ${CALENDLY_TOKEN}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      event_type: CALENDLY_EVENT_TYPE_URI,
      start_time: slot.start,
      invitee: {
        name: `${leadData.firstName} ${leadData.lastName}`,
        email: leadData.email,
        text_reminder_number: leadData.phone,
      },
      questions_and_answers: [
        {
          question: "What would you like to discuss?",
          answer: leadData.interest,
        },
        { question: "Budget range", answer: leadData.budget },
      ],
    }),
  });

  return response.json();
};

GoHighLevel Calendar:

const bookGHLAppointment = async (calendarId, slot, leadData) => {
  const response = await fetch(
    "https://rest.gohighlevel.com/v1/appointments/",
    {
      method: "POST",
      headers: {
        Authorization: `Bearer ${GHL_API_KEY}`,
        "Content-Type": "application/json",
      },
      body: JSON.stringify({
        calendarId,
        selectedSlot: slot.start,
        selectedTimezone: leadData.timezone || "America/New_York",
        contact: {
          id: leadData.contactId,
        },
        title: `Call with ${leadData.firstName}`,
        meetingLocationType: "phone",
        appointmentStatus: "confirmed",
      }),
    },
  );

  return response.json();
};

Step 5: Objection Handling

const objectionHandlers = {
  price: {
    detect: ["expensive", "cost", "price", "afford", "budget"],
    response: `I totally understand wanting to make sure it's worth it. What would the ROI need to look like for it to make sense? That way I can make sure we cover that on the call.`,
  },
  timing: {
    detect: ["busy", "not a good time", "swamped", "later"],
    response: `No worries - I can book something a few weeks out if that's easier? Or if you'd prefer, I can follow up next month.`,
  },
  competitor: {
    detect: ["already working with", "have someone", "competitor"],
    response: `Got it! If you ever want a second opinion or comparison, we're always here. Is there anything specific you wish was better with your current setup?`,
  },
  thinkAboutIt: {
    detect: ["think about it", "consider", "discuss with"],
    response: `Totally fair! What questions do you still have that I could help answer? Sometimes that helps with the decision.`,
  },
  notReady: {
    detect: ["not ready", "just looking", "researching"],
    response: `No problem at all. Would it be helpful if I sent over some info you could review on your own time? What would be most useful?`,
  },
};

const detectObjection = (message) => {
  const lowerMessage = message.toLowerCase();

  for (const [type, handler] of Object.entries(objectionHandlers)) {
    if (handler.detect.some((phrase) => lowerMessage.includes(phrase))) {
      return { type, response: handler.response };
    }
  }

  return null;
};

Step 6: Human Handoff

const shouldHandoff = (conversation, leadScore) => {
  const reasons = [];

  // High-value lead
  if (leadScore >= 90) {
    reasons.push("high_value_lead");
  }

  // Requested human
  const wantsHuman = [
    "speak to someone",
    "talk to a person",
    "real person",
    "human",
  ];
  if (
    wantsHuman.some((phrase) =>
      conversation.some((msg) => msg.content.toLowerCase().includes(phrase)),
    )
  ) {
    reasons.push("requested_human");
  }

  // Complex question AI can't handle
  if (conversation.length > 10 && !hasBooked) {
    reasons.push("extended_conversation");
  }

  return reasons.length > 0 ? reasons : null;
};

const executeHandoff = async (contactId, reasons) => {
  // Alert sales team
  await sendSlackAlert({
    channel: "#sales-handoffs",
    message: `Human handoff requested for ${contactName}`,
    reasons,
    conversationLink: getConversationLink(contactId),
  });

  // Update contact
  await updateContact(contactId, {
    tags: ["needs-human-follow-up"],
    assignedUser: getNextAvailableRep(),
  });

  // Send message to lead
  await sendSMS(
    phone,
    `Great question! Let me get someone who can help with that. ` +
      `${repName} will reach out shortly - usually within the hour!`,
  );
};

Complete n8n Workflow

1. Webhook: Receive inbound SMS
      ↓
2. Get Contact from GHL (by phone number)
      ↓
3. Get Conversation History (from custom field)
      ↓
4. Check for Objections → Handle if found
      ↓
5. Check Qualification Status
      ↓
6. Generate AI Response (with context)
      ↓
7. Check for Booking Intent → Fetch calendar if yes
      ↓
8. Send Response via Twilio
      ↓
9. Update Conversation History in GHL
      ↓
10. Check if Qualified & Booked → Trigger next workflow

Sample Conversations

Successful Booking

AI: Hey Sarah! Thanks for reaching out about our marketing services. What specifically are you looking to improve?

Lead: We need more leads for our HVAC business

AI: HVAC - love it! We work with a lot of home services companies. Quick question - are you looking for local leads, or do you service a wider area?

Lead: Local, about 30 mile radius

AI: Perfect. And roughly what's your current monthly budget for marketing?

Lead: We can do about 3-4k

AI: That's a great starting point for local lead gen. Last question - are you the one who handles marketing decisions, or would anyone else be involved?

Lead: It's my company, I decide

AI: Awesome. I'd love to get you on a call with our HVAC specialist - he's helped companies like yours double their lead flow. I have openings:

1. Tomorrow at 2pm EST
2. Thursday at 10am EST
3. Friday at 3pm EST

Which works best?

Lead: Thursday 10am

AI: Booked for Thursday at 10am EST. You'll get a calendar invite and reminder. Is this the best number to call you at?

Lead: Yes

AI: You're all set. Mike will call you Thursday at 10am. He'll have some questions about your current marketing and show you what's worked for other HVAC companies. Talk soon!

Objection Recovery

AI: Hi! This is AI assistant from [Company]. You filled out our form about sales training - what prompted you to look into this?

Lead: Our team isn't hitting quota consistently

AI: That's frustrating. How many reps are on the team?

Lead: About 15

AI: Got it. And is this something the company has budget allocated for, or are you still building the case?

Lead: Honestly we probably can't afford it right now

AI: I hear you - budget's always tight. Out of curiosity, what do you think missing quota is costing you monthly? Sometimes that helps put training investment in perspective.

Lead: Good point. Probably 50k+ in lost revenue

AI: Ouch. Our program typically costs about 10% of that and clients see results within 90 days. Would it make sense to at least see what the program looks like, so you have the info when budget frees up?

Lead: Yeah that would be helpful

AI: Great! I can get you 30 minutes with our training director. He'll show you exactly what the program covers. How's Thursday at 2pm?

Metrics to Track

---

Ready to automate your appointment setting? Get the implementation package or book a call for done-for-you setup.