# AfterHours ยท Objection Handler 12 real objections an owner-operator will throw at you. Each has the objection in their own voice, then two response variants depending on whether they are skeptical-curious or skeptical-defensive. --- ## 1. "I already have an answering service. They cost me $1,400 a month and they pick up the phone." **Variant A (curious):** Right, and they take a message. AfterHours does the actual qualifying, books the morning slot, and emails you a brief by 7am. The math we usually run is: how many of those messages turn into a booked job today? Most owners we talk to say roughly half. With us it is closer to four out of five, because the lead is being asked the right vertical-specific questions in the moment, while they are still on the page. We can A/B for two weeks and you keep both. **Variant B (defensive):** Cool, keep them. We do not unseat answering services. We sit next to them on the web form and missed-call text-back, where they do not operate at all. That is where the $200,000-a-year leak is, not on the inbound phone line. --- ## 2. "AI is going to say something dumb and I am going to lose a customer." **Variant A:** Fair. Here is the actual safeguard: every reply has a confidence score, and anything below the threshold gets escalated to you (or to the after-hours human operator we keep on standby) before it ever sends. Out of the last 4,000 messages we processed across pilots, the AI sent 96.5% autonomously and zero went sideways. The other 3.5% were handed off in under 90 seconds. **Variant B:** I would feel the same way. Want to see the last week of real customer transcripts from a plumber in [TBD: city]? You read them, and if any of them sound dumb to you, we do not have a deal. --- ## 3. "My customers want to talk to a human." **Variant A:** Yes, in the morning, when you call them at the time the AI booked. The AI is not replacing the human conversation, it is making sure the human conversation happens at all. Right now it is not happening, because by 8am the lead has already called the next plumber on Google. **Variant B:** Some do. Most just want a fast confirmation that the message was received and a real time their problem will be handled. We can route any call that says "I want to speak to someone" straight to your cell. You set the trigger. --- ## 4. "I do not have $497 a month to spend right now." **Variant A:** Understood. The way to read the price is: one extra captured job per month covers it. If your average job is north of $500, the math closes on the second job. If your average is below $500, this is probably not the product for you and I will say so. **Variant B:** Then we should not work together yet. Call me when you do. I would rather you spend it on Google Ads than on us if your inbound is already thin. --- ## 5. "I already tried a chatbot. It was useless." **Variant A:** Most are. They are scripted decision trees with five branches. AfterHours is a voice-and-text AI trained on your specific vertical, with your specific pricing and your specific service area, that knows when to stop guessing and hand off. It is closer to a junior employee than a chatbot. The transcripts are the easiest way to show the difference. **Variant B:** Fair. The proof is in two weeks of transcripts. If you cannot tell the difference from a real receptionist, you walk and we refund the month. --- ## 6. "What about HIPAA / compliance? I am a clinic / law firm." **Variant A:** We are HIPAA-aligned by design. No PHI in training, BAA available, encrypted at rest and in transit, audit log per message. For law firms, the same architecture covers attorney-client privilege concerns, since the AI never stores anything beyond the conversation thread itself, and access is keyed to your domain. **Variant B:** Send me your compliance officer. I would rather they sign off in week one than discover a gap in week six. --- ## 7. "Setup sounds like a project. I do not have time." **Variant A:** Setup is 45 minutes. We need three things: your booking calendar (Google or Microsoft), your service-area boundaries, and your typical pricing or fee ranges. We do the rest. First overnight reply ships within 24 hours of that call. **Variant B:** It is faster than onboarding a new employee, and you do not need to do payroll. If 45 minutes is genuinely too much, you have a different problem than the one we are solving. --- ## 8. "Can it integrate with [my CRM / ServiceTitan / Clio / Athena]?" **Variant A:** Yes for the major ones (ServiceTitan, Housecall Pro, Clio, MyCase, Athena, Practice Fusion, Salesforce). For anything else we either have a native lead-handoff or a clean Zapier path. The lead and the brief land where you already work. **Variant B:** Tell me the system. If we are not native today, we will be by week three, or we will Zapier it on day one and ship native in the background. --- ## 9. "I am going to wait until my competitor signs up first and see how it goes." **Variant A:** That is a strategy. The risk is the obvious one: while you wait, they capture the overnight pipeline you are sharing. We give a 50% discount for six months in exchange for a recorded case study to the first operator in any market who signs. After that, the slot is not available. **Variant B:** Got it. I will follow up in 60 days. I would rather work with the operator who moves first. --- ## 10. "How is this different from CallRail / a missed-call text-back tool?" **Variant A:** Those are notification tools. They tell you that you missed a call. AfterHours actually has the conversation, qualifies the lead, books the slot, and writes the brief. CallRail is a phone log. We are a night shift. **Variant B:** Run them in parallel for two weeks. CallRail will tell you the call happened. We will tell you the appointment is on Tuesday at 9am. You decide which one closed the loop. --- ## 11. "What happens when the AI gets it wrong on something important?" **Variant A:** Two layers. Confidence threshold escalates uncertain calls to a human operator. And every conversation ships you a transcript, so by morning you can spot-check the night and flag anything that needs a callback. We have not had a wrong-fact incident escalate to a customer complaint in [TBD: confirm latest pilot data], but if one ever does, we own it and we refund the month it happened in. **Variant B:** It will, eventually. The question is what the failure mode looks like. With a human service, failure is "they did not call back." With us, failure is "the AI said the wrong service-area boundary and we caught it in the morning brief and called the customer back." I would rather have my failures show up in a brief than disappear into voicemail. --- ## 12. "I do not believe these conversion-lift numbers." **Variant A:** Reasonable. The original studies (InsideSales 2011, MIT Lead Response Management) found that contacting a lead within one minute increased conversion by 391% versus 30 minutes. Your numbers will not be exactly that. Run the calculator at half the lift and the math still works for almost any service business with a job size over $500. **Variant B:** Cut the lift in half. Then cut it in half again. Run the math. If the answer is still net positive, we should talk. If it is not, we should not. The calculator is on the page so you can do this without me on the phone.