The difference between a good and bad appliance repair shop usually shows up before the truck rolls. A great intake captures the appliance brand, the model number, the serial, the symptom in the customer's own words, the age, the recent repair history, and the parts the tech is going to need. A bad intake captures "the fridge isn't cold" and an address.
Most answering services, live or AI, do the bad intake by default. The reason is that the intake schema was built for plumbing or HVAC or a generic trade, and the appliance-repair-specific fields were never added. This post walks through what good intake looks like, what each field changes about the tech's day, and how to set this up if you're piloting an answering service for the first time.
Why intake quality drives margin
Appliance repair runs on first-trip fix rate. Every truck roll where the tech doesn't have the right part is a margin-eater: re-trip cost, parts pickup, customer scheduling friction, and the higher likelihood of cancellation when the customer waits an extra week.
The intake fields that drive first-trip fix rate are not negotiable. If your answering service is capturing "name, phone, address, appliance type", you are running the worst-case intake. Let me show what the good version looks like.
The minimum viable intake schema for appliance repair
Eight fields. Every call. No exceptions.
- Customer name and best callback number.
- Service address (with apartment/unit number if applicable).
- Appliance type. Fridge, dishwasher, oven, range, microwave, dryer, washer, disposal, etc.
- Brand. Whirlpool, GE, LG, Samsung, Bosch, Sub-Zero, Wolf, etc.
- Model number. Critical. Without this, the tech can't pre-pull parts. The script should walk the customer through where to find the model plate on each appliance type (inside the fridge door, on the dishwasher kickplate, on the back of the oven door, etc.).
- Symptom in the customer's own words. "It's not cold" vs "it's running but the freezer is warm" tell different stories.
- When the symptom started. Today, last week, gradual over a month.
- Has this been worked on before? By you or anyone else, and what was done.
The model number field is the single highest-impact addition to most appliance repair intakes. A tech who knows the model can look up the parts diagram, the common failure modes, and the typical first-trip parts before leaving the shop. That alone moves first-trip fix rate from 60% to 85%+.
The safety branches that should fire first
Like any trade-specific dispatcher, appliance repair has a small number of intents that should preempt the standard intake.
Gas range pilot out with gas smell: Stop the intake. Tell the customer to turn off the appliance shutoff valve if they can find it safely, ventilate the room, and call the gas company if the smell persists. Do not schedule until gas company has confirmed safe.
Refrigerator down with insulin or temperature-sensitive medication: Same-day priority. Confirm the medication is still cold and give the customer the bridge advice (move to a cooler with ice; check pharmacy refill options). Dispatch tech as soon as possible.
Dryer with burning smell: Stop the intake. Tell the customer to unplug the dryer and not run it again. Schedule same-day inspection. Dryer fires are not theoretical, lint fires are real.
Washer leak through a ceiling: Stop the intake. Tell the customer to shut off the water supply to the washer if they can find it safely. Confirm the leak is contained. Dispatch tech same-day, or refer to a plumber if the customer doesn't have one.
Freezer down with breast milk or large food perishables: Sympathetic same-day priority. Give the customer the bridge advice and book an urgent slot.
These branches are why a generic answering service often fails appliance repair shops. The branches aren't in the script unless someone built them in.
OnCrew does not carry an appliance branch list: appliance repair is not one of its four built-in trade profiles (HVAC, plumbing, electrical and roofing). On an OnCrew line a call is urgent when the caller says it is or describes something that cannot wait, and if anyone is in immediate danger (fire, smoke, a medical emergency) Emily tells the caller to call 911 first. The bridge advice above belongs to your on-call tech's callback.
The dispatch tags that move calls through the queue
Once the intake is captured, the dispatch tag determines what happens next. The four-tag model:
| Tag | Definition | Routing |
|---|---|---|
| EMERGENCY | Safety branch fired (gas, burning, active leak, refrigerator down with insulin) | Same-day, immediate handoff to on-call |
| URGENT | High-value appliance offline, large food at risk, primary appliance down (only fridge, only oven) | Same-day or next-day priority slot |
| ROUTINE | Functional issue, customer can work around it (ice maker not making ice, microwave display dim) | Next available slot, 2-5 days |
| ESTIMATE | New install, upgrade, replacement quote | Sales/estimating queue |
A good answering service tags every call before it hits your dispatcher. A bad one dumps everything in one queue. On OnCrew each call arrives marked urgent or routine, with its summary and transcript, and your dispatcher applies the four tags from there.
Where the model number capture breaks
If you've ever asked an answering service to capture model numbers, you've seen the failure mode: the customer doesn't know where the model plate is, the agent doesn't know either, and the call ends without it.
The script should walk the customer through the model plate location for each appliance type. A good receptionist, live or AI, can do this conversationally:
- "For a fridge, the model plate is usually on the upper-left wall inside the fresh food compartment, or on the kickplate at the bottom front. Can you check?"
- "For a dishwasher, it's on the inside edge of the door, around the latch. Can you open the door and look at the side or the top edge?"
- "For an oven, it's usually on the frame inside the door or on the door itself. Open the oven door and look around the frame."
If the customer can't find it, the fallback is to capture the appliance description in enough detail (size, color, age, configuration) for the tech to do the lookup on arrival. Don't make perfect the enemy of good.
What changes for the tech with good intake
A tech walking into a fridge call with full intake:
- Model number → pulled the most likely diagnostic parts (start relay, evaporator fan, condenser fan motor, defrost thermostat) before leaving the shop.
- Symptom + when it started → already has a working theory.
- Repair history → knows whether it's a warranty job, a callback, or a fresh failure.
- Age → knows whether to push for replacement or invest in repair.
A tech walking into a fridge call with bad intake:
- "Not cold" → could be 12 different things.
- No model number → can't pre-pull anything.
- No history → could be a callback on someone else's work.
The first tech wraps the call in 90 minutes with parts in hand. The second tech wraps in 2 hours, no parts, comes back tomorrow with the right parts after the supply house opens. Same fault, same labor, same parts cost. Different revenue per hour.
How to test your current answering service
Pick five recent jobs where the tech needed a re-trip because the part wasn't on the truck. Pull the intake notes for those five calls. Score them:
- Was the model number captured?
- Was the symptom captured in the customer's words or paraphrased into a category?
- Was repair history asked?
- Was age captured?
If three of the five are missing model numbers and history, your intake is the problem. Switching to better intake, whether AI or live, will recover margin you didn't know you were leaking.
For deeper reading, see the appliance repair call intake guide, the appliance repair answering service page, and the AI receptionist product page.
FAQs
How long does the average appliance repair intake call run?
With full schema capture, 3-5 minutes on a routine call, 4-7 on an emergency. The model-number capture adds 30-60 seconds. The trade-off can be worth it when a slightly longer intake helps avoid repeat callbacks.
Can the AI receptionist look up the parts diagram during the call?
No. The diagram lookup is a dispatcher or tech task. What an AI receptionist can do is ask for the model number on routine calls (on OnCrew, when your notes list it) so the dispatcher does the lookup before assignment.
What about commercial appliance repair?
The schema is mostly the same with two additions: capture the establishment name (restaurant, hotel, laundromat) and the contact role (general manager, kitchen lead, owner). Commercial calls also need a "kitchen down" tag that moves to same-hour priority, a restaurant with no cooking equipment is losing thousands of dollars an hour.
How do I migrate intake quality from voicemail to a live or AI service?
Run both for two weeks. Listen to the new intakes. Spot-check against your tech's "did you have the right parts?" rate. The metric you're moving is first-trip fix rate; that's the proof.