Use cases / Customer Service

Set up an AI receptionist for missed calls

Editor-reviewed

Summary

You're in the field most of the day, so calls go to voicemail; missed calls can mean missed jobs, and routine questions about services and availability eat time when you do pick up. Short answer: use Goodcall (standalone 24/7 AI phone agent if you do not run field service software; vendor claims, not tested by us). Free option: ChatGPT. The workflow has 8 steps and 3 copy-paste prompt(s); expect a one-page FAQ and rule sheet you can paste into the receptionist's settings, a test-call checklist, and a weekly review routine. How well it works depends on the vendor and your setup. Main caution: Vendor descriptions are the vendor's own claims; we have not tested the product.

The problem

You're in the field most of the day, so calls go to voicemail; missed calls can mean missed jobs, and routine questions about services and availability eat time when you do pick up.

All tools mentioned

Step-by-step workflow

  1. List the questions callers ask most (services, service area, whether you do free estimates, how soon you can start) and write the answer you would give to each.
  2. Write the rules: what the AI may do (take a message, create a request, book an estimate visit) and what it must never do (quote unapproved prices, promise start dates, handle complaints or billing disputes).
  3. Pick escalation triggers that should reach you right away (for example an irrigation leak, a tree down, or a commercial account name). Jobber and Housecall Pro both describe sending calls to a person when needed.
  4. Choose the tool. If you already run Jobber or Housecall Pro, their receptionist add-ons use the services and booking settings already in your account, per their own pages. Otherwise compare a standalone AI phone agent such as Goodcall.
  5. Keep your existing number: both vendors describe forwarding calls from your current carrier; Jobber describes forwarding after hours or after a set number of rings.
  6. Start narrow: have it answer only when you can't (after hours, during jobs), and limit bookings to your service area and real booking windows.
  7. Test it yourself before going live: call as a vague customer, an out-of-area customer and an urgent one, and fix every wrong answer. Jobber offers a demo line for this.
  8. For the first few weeks, read every call summary or transcript and call back anyone the AI could not help.

Copy-paste prompts

Prompt 1
I run a [maintenance / design-build / irrigation] landscaping company serving [towns or ZIPs]. Services: [list]. We don't do: [list]. Minimum job: [ ]. How estimates work: [ ]. Hours: [ ]. Write the 12 questions callers most likely ask and a short, plain answer to each using ONLY these facts. Mark any question you can't answer from these facts as TAKE A MESSAGE.
Prompt 2
Write a one-page rule sheet for an AI receptionist at my landscaping company: what it may do [take messages / create requests / book estimate visits], what it must never do (quote prices, promise start dates, discuss complaints or billing disputes), phrases that should transfer the call or text me immediately [list], and a polite line for callers outside our service area [area].
Prompt 3
Here are this week's AI receptionist call summaries [paste]. List: calls it handled correctly, calls where its answer was wrong or incomplete (quote the line), callers who still need a callback, and changes I should make to its FAQ or rules.

What to expect

A one-page FAQ and rule sheet you can paste into the receptionist's settings, a test-call checklist, and a weekly review routine. How well it works depends on the vendor and your setup.

Limits & safety

  • Vendor descriptions are the vendor's own claims; we have not tested the product.
  • AI can get things wrong: Housecall Pro's own help center says CSR AI can occasionally make mistakes even with the best setup. Review summaries and call people back.
  • Some callers prefer a person, especially for complex or custom projects; keep an easy way to reach you.
  • Check your state's rules on recording calls and disclosing automated systems before you turn it on, and set the greeting to match.
  • Booked-job and time-saved figures posted by AI agencies are their own self-reported marketing claims.

Source / inspired by

Free weekly email: 3 new use cases

One short, plain email each week with three practical AI use cases for landscaping companies. Free; unsubscribe anytime. Read a sample issue →

Subscribe free on Beehiiv →Signup is handled by Beehiiv. We only use your email to send the weekly issue.
Sponsors
dailywork.aiThe operating system for your landscape companyOpen sponsor slotSponsor this site — reach landscaping business ownersOpen sponsor slotSponsor this site — reach landscaping business owners