How to Sort New Customer Inquiries by Job Type and Urgency With AI

When inquiries arrive from email, web forms, voicemail, and text, everything feels urgent. It is easy to spend time on low-fit leads while higher-value jobs sit unanswered.

Who this is for

This article is for small business owners and operators who handle their own intake and need a fast way to prioritize who gets a reply first.

Prerequisites

• A batch of recent customer inquiries pasted into one document or spreadsheet.
• An AI tool such as ChatGPT, Gemini, or Claude.
• A simple prioritization rule such as emergency, same-day quote, normal quote, referral out, or not a fit.

Step-by-step workflow

1. Build a simple intake taxonomy

Decide your categories before you ask AI to sort anything. Keep them small and operational.

2. Paste in a batch of inquiries

You can paste multiple messages at once and ask the model to classify each inquiry by job type, urgency, and quote readiness.

3. Ask for a table, not a paragraph

A table makes it easier to review and copy into a tracker.

4. Add a recommended next action

Have the model suggest one next action such as quote now, ask two missing questions, call today, or decline politely.

5. Review edge cases yourself

AI can accelerate triage, but you should still review anything high-value, unusual, or ambiguous.

Tool-specific instructions

ChatGPT and Gemini are both good when you want to paste or upload batches of inquiries and ask for structured outputs. Claude is useful when you want especially clean writing in the follow-up actions and templates that come after triage.

Copy/paste prompt block

{
  "task": "Classify customer inquiries for a small service business",
  "instructions": [
    "Read every inquiry in the batch.",
    "For each one, assign: job type, urgency level, fit level, and quote readiness.",
    "Use these urgency labels only: emergency, same-day, routine, low-priority.",
    "Use these fit labels only: strong fit, possible fit, not a fit.",
    "Suggest one next action per inquiry.",
    "Return the result as a table with columns: inquiry_id, customer_name_if_known, job_type, urgency, fit, missing_details, next_action."
  ],
  "input": "Paste the inquiry batch here."
}

Quality checks

Spot-check at least five rows. Make sure urgent messages were not misread as routine, and make sure not-a-fit leads are genuinely outside your area, service type, or minimum job size.

Common failure modes and fixes

The model can over-prioritize emotionally worded inquiries even when they are not profitable or in scope. Fix that by telling it exactly what counts as emergency and what counts as a strong fit in your business.