For a phone repair shop, start with automatic status texts from your repair-ticket software, which handle most "is it ready?" calls without AI. Add an AI receptionist or chatbot only for the calls that remain, and only if it can read ticket status or hand over cleanly. Monthly costs range from nothing extra to a few hundred dollars.
The reason for starting with templates is volume. A customer rings because they were not told something: that the part arrived, that the quote is ready, that the phone can be collected. Fix those four moments and most calls disappear. Those messages come first below, then a comparison of text templates, chatbots and AI reception on cost and risk.
The four moments that cause most calls
Look at your last fifty calls and sort them. Typical repair-shop calls fall into a few groups:
- "Is my phone ready?" (status)
- "How much will it cost?" (quote)
- "Can you fix X, and how long?" (new enquiry)
- "What time do you close?" (information)
- "I'm outside, where do I collect it?" (pick-up)
The first and the last are the ones a status text prevents. The second is a quote message, and the third and fourth are enquiry handling. Count your own, because a shop with a lot of walk-in trade and a shop that takes mail-in repairs have different mixes. For intake itself, see how to turn customer messages into tickets with AI.
Status texts: the four that matter, with wording
Most repair-shop systems can send an automatic text or email when a ticket changes status. RepairDesk's page, for example, describes automatic job updates by text and email, and RepairShopr's lists ticket progress tracking and a customer web portal; neither page gave me prices. Check that your current software does the same before buying anything new. Four messages cover the journey:
- Received. "Hi [first name], we've got your [device] and logged repair #[number]. We'll text you a quote within [time]. Reply here with any questions. [shop name]"
- Quote ready. "Hi [first name], your quote for the [device] is $[amount] for [repair], taking about [time]. Reply YES to go ahead or call us on [number]."
- Complete. "Good news, your [device] is ready for collection. We're open [hours]. Please bring [ID or ticket number]. Balance due: $[amount]."
- Reminder. "A quick reminder that your [device] is ready to collect. We keep devices for [number] days, so please collect by [date]."
Two details raise the value of these. The quote message should state what happens if the customer says no (a diagnostic fee, return of the phone), because unclear quotes create calls and disputes. And the "complete" message should tell them what to bring, because "I forgot my receipt" is a five-minute counter conversation.
Before and after: the "complete" text
Before: "Your order is ready." The customer calls to ask which order, whether payment is needed, and when you close.
After: "Hi [first name], your iPhone screen repair #4821 is ready. We're open 9:00-18:00 Mon-Sat. Balance $79, card or cash. Bring this message or photo ID." The second version answers three questions at once and carries the ticket number, so an assistant or a colleague can find it fast. Use a placeholder in the template and let the system fill the real values.
Three types of tool, compared on the repair job
| Option | Handles | Cannot handle | Cost shape |
|---|---|---|---|
| Status text templates in repair software | Received, quote, complete and reminder messages; reduces most "is it ready?" calls | Open questions and phone calls | Often included with the repair software; check your plan |
| Website or chat chatbot | Opening hours, price ranges, booking, simple FAQ; some can look up an order if connected | Real ticket status unless integrated; phone calls | Free tiers exist (for example, Tidio's free plan includes 50 free AI conversations that do not renew); paid plans rise with volume |
| AI reception (phone) | Answers calls, takes a message, sends a text, may book a drop-off | Judging a diagnosis; ticket status unless integrated | Per call or per customer helped; examples below |
The point of the table is that none of the three replaces the first. A call asking "is it ready?" has one correct answer, and it sits in your ticket software. An AI receptionist that cannot read the ticket can only take a message, which is better than a missed call but no better than a text that told the customer already.
What AI reception costs, as listed
Prices change, so take these as a snapshot and check the live page. As of the pages I fetched:
- Goodcall says plans start at $79 a month (Starter, 100 unique customers included, then 79 cents each beyond that), billed by unique customers rather than call length, with no per-minute charges. It texts you lead details at call end and can text booked appointments to the people involved.
- Smith.ai (fact sheet, September 2026): its AI Receptionist is free for 25 calls, then $150 for 75; its human plans start at $300 for 30 calls.
- RepairDesk describes a backup AI receptionist (ARIA) that answers missed calls and books repairs; its page lists no price, so ask.
- Intercom's Fin charges $0.99 per resolved outcome, which suits shops with a busy website chat and is probably more than a small repair counter needs.
A monthly cost sum for a shop with 300 calls
Say an illustrative shop takes 300 calls a month, of which 120 are "is it ready?" and 60 are new enquiries; the rest are collections, prices and hours. Suppose status texts remove two thirds of the status calls. That leaves 40 status calls plus 180 others, or 220 calls.
If an AI receptionist answers only the calls you miss while at the bench, say a third, that is 73 calls. On a plan of $150 for 75 calls (the Smith.ai AI Receptionist rate above), the cost is $150 a month. If you handle the same calls yourself, each takes about three minutes, so 73 calls cost 3.6 hours. At $20 an hour that is $73. The AI is dearer than the labour on those numbers, and worthwhile only if the missed calls would otherwise turn into lost repairs. If one lost screen repair is worth $90 of margin, then recovering two a month covers the $150.
That is the honest decision rule: AI reception pays when it converts or saves calls that carry money, not when it answers questions a text could answer. For a longer cost treatment in a neighbouring trade, see what phone answering costs a small repair shop.
Teaching a receptionist or chatbot to handle your calls
Whatever tool you choose, it needs the same facts, written once. Prepare a one-page brief:
Shop: [name], hours Mon-Sat 9:00-18:00, closed Sunday.
Repairs we do: phone screens, batteries, charging ports, back glass,
water-damage checks. We do not repair: [list].
Price ranges (guide only, final quote after inspection): screen
$[x]-$[y], battery $[x]-$[y]. Diagnostic fee $[amount], deducted if
you go ahead.
Typical times: screens same day if in stock, otherwise 2-3 days.
Status: never state a repair status; say a text will be sent, and
offer to take a message for the technician.
Collection: bring the ticket number or photo ID.
Hand over to a person: data recovery, business accounts, complaints.The line about status is the one that prevents harm. An AI that confidently says "your phone is ready" when it is not sends a customer across town for nothing.
Test calls to run before going live
- "Is my phone ready? It's under Sam." (It must not guess; it should take a message.)
- "How much is an iPhone screen?" (A range with "final quote after inspection".)
- "My phone fell in water, can you save my photos?" (Hand-over to a person; no promise.)
- "I want a refund, the screen came off." (Hand-over; apology without liability admission.)
- A noisy call, an accent, a bad line. (Does it ask to repeat, or invent?)
Record each result, and keep the failures as a permanent test list. If you take payments or customers' data over the phone, check what the tool stores, and tell callers that a call may be recorded or answered by an AI where that is required. Our tutorial on setting up an AI phone agent for after-hours calls covers the set-up steps in detail.
Quote replies: the one place AI drafting helps
Quote enquiries arrive by message at all hours: "How much for an iPhone 13 screen?" A tool or a template can answer them, with a range and a next step. Give an assistant your price table and ask it to draft a reply per enquiry that you approve, or set a template.
Reply to this enquiry for a phone repair shop using only the price
table below. Give the price range, the time, and one question that
moves the customer to booking. If the device or fault isn't in the
table, say we'll confirm after inspection. Under 60 words.
Enquiry: "Cracked screen on my Galaxy S22, how much?"
Price table: [paste]Illustrative reply: "Hi [first name], a Galaxy S22 screen is usually $[x]-$[y] and takes about [time] if the part is in stock. Final price after a quick check. Want to bring it in today or tomorrow?" If a quote goes unanswered, a follow-up two days later recovers some of them, which following up repair estimates with AI explains.
Pick-up reminders that reduce abandoned devices
Uncollected phones take up shelf space and cause worry. A reminder schedule that works for most shops is: a "ready" text on completion, a reminder at three days, and a final notice at the point your storage policy applies, which you should state on the ticket. Keep the final notice factual: it names the date, the storage terms, and the contact number. Check what your local rules say about unclaimed goods before you write a policy that includes disposal or sale, and get that wording confirmed by an adviser.
One last sum shows why reminders matter. If a shop has 12 devices a month waiting more than a week for collection and each holds $60 of unpaid balance, that is $720 of money sitting on a shelf. A reminder at three days that brings back even a third of them recovers $240 a month, more than most tools in this tutorial cost. Check your own shelf before you decide what to buy.
A week of messages for one shop
An illustrative shop handles 25 repairs a week. With four automatic texts per repair, that is 100 messages a week, all produced by the repair software. If each message prevents even a third of a call, that is 33 calls avoided, at three minutes each: 99 minutes a week back at the bench. Add one reminder for the 5 devices still waiting after three days. The sum is rough, but it explains why the template layer comes first: it costs the least and removes the most calls. If your call count does not fall within a month, check that messages are being sent on every status change, which is the commonest fault.
Status wording for the awkward cases
The four standard texts cover a smooth repair. Customers call most when something goes off script, so write these three too and keep them ready in your system:
- Part delayed. "Hi [first name], the [part] for your [device] hasn't arrived, so we now expect it on [date]. We'll text as soon as it's in. If you'd rather collect the phone as it is, reply COLLECT and we'll have it ready." Offering the option reduces the angry call.
- Extra fault found. "While repairing your [device] we found [fault]. Fixing it would add $[amount]. We haven't done any extra work. Reply YES to go ahead or NO to collect it as quoted." Never carry out extra work on an unanswered message.
- Cannot repair. "We've checked your [device] and can't repair it economically because [reason]. There's no charge beyond the $[amount] diagnostic fee. It's ready to collect, and we can talk through options."
These messages are where a human reads the draft before sending, even if an assistant wrote it. Money and bad news are not the place for an unreviewed automatic reply.
A Monday at the counter, with and without messages
An illustrative Monday: the shop opens at 9:00 with eleven devices on the bench. Without status texts, the phone rings fourteen times before noon, nine of them "is it ready?". Each call interrupts a repair, and a screen repair that needs steady hands takes longer. With status texts, five customers received "ready" messages at their completion, and the shop gets perhaps four status calls, all from people who did not read the text. The technician finishes the morning with two more repairs done. These are example figures, but the mechanism is real: every interruption costs more than the call itself.
To measure your own, keep a tally sheet by the phone for a week, with a column for each call type. Repeat it a month after you switch on the texts. A drop in the "is it ready?" column is the proof you need, and a column that has not moved tells you the texts are not being sent or read.
Audit your ticket statuses first
Automatic texts are only as good as the statuses behind them. List your statuses and what each triggers. An illustrative audit for a small shop:
| Status in the system | Triggers a text? | Fix needed |
|---|---|---|
| Received | Yes | Add expected quote time |
| Diagnosing | No | Fine; no text needed |
| Awaiting approval | Yes (quote) | Add "what if I say no" |
| Awaiting part | No | Add delayed-part text |
| Repaired | No | Merge into Ready, or the customer is never told |
| Ready for collection | Yes | Add hours and what to bring |
The "Repaired" row is the common gap: technicians finish the job but the status that sends the text is a different one, set later by the counter. Merge them, or make setting the status part of the technician's last step.
Mail-in repairs and customers who live far away
If you take repairs by post, status messages carry more weight, because the customer cannot walk in. Add a text when the parcel arrives, another when the repair starts, and one with the tracking number when it ships back. Customers who send a device away are anxious, and a short message every step removes most "where is it?" contacts. Where an enquiry arrives by social message rather than a call, our guide to one AI inbox for WhatsApp, Instagram and Facebook enquiries shows how to bring them into one place.
Privacy and passcodes
A phone holds a customer's photos, messages and banking. Keep these rules in your messages and any AI tool:
- Never ask for a passcode over text or chat. If a repair needs one, collect it on paper at the counter and destroy it at collection.
- Do not paste customer messages with names and numbers into a consumer chat assistant; use placeholders, or a tool with a business agreement.
- Check where an AI reception service stores call recordings, and for how long.
- Tell callers when a call is recorded or answered by an AI, wherever your rules require it.
For wider questions about AI tools and shop data, our guide to logging call summaries in your CRM shows how to keep a record of calls without storing more than you need.
Where each tool trips up
- Templates: staff forget to move a ticket to "complete", so the text never sends. Add a daily check of tickets that have sat in the same status for three days.
- Chatbots: an out-of-date price list. Review the answers monthly against the real prices.
- AI reception: misheard device models and names. Make it repeat back the model and a phone number.
- All of them: a customer who wants a person gets stuck in a loop. Always offer "reply CALL and we'll ring you".
Choose in this order: status texts first, then a chatbot or AI reception only for what remains, and a person for anything involving data recovery, complaints or money. If you can only do one thing this month, switch on the "ready for collection" text and read the next week's call log.
Questions repair shop owners ask about support tools
Is a chatbot enough for repair status questions?
A chatbot on your website helps with opening hours, prices and booking. It cannot answer a repair status unless it can read your ticket system. If most questions are 'is it ready?', fix the status text first, because a clear automatic message removes most of those calls before any AI is needed.
Can AI answer the phone when we are busy at the bench?
Yes, AI reception services can answer, take a message, send a text and sometimes book. Test whether it can give a real ticket status. If it cannot read your system, it should take a message and promise a text back, not guess.
How many status messages should a repair send?
Four is usually enough: received, quote ready, repair complete and ready for pick-up, with one reminder after a few days. More than that trains customers to ignore your messages. Send extras only when something changes, such as a delayed part.
What does this cost per month for a small shop?
Text templates inside your repair software can cost nothing extra. AI reception and chatbot plans vary widely and are often billed per call or conversation. Work out your monthly call count, multiply by the vendor's per-call rate, and compare with the hours your team spends answering calls.
Further reads
- How Garages Use AI to Send Repair Updates Customers Appreciate — How garages send repair updates customers appreciate.
- Phone Repair Shop Marketing: Local Search and Trade-In Offers With AI — Bring in more repair and trade-in customers.
- Shoe Repair Software: Job Tickets, Pick-Up Texts and AI Replies — The ticket fields, quote texts and unclaimed-item wording a shoe repair shop needs, and where AI can draft the replies.
- AI for a Jewellery Business: Anniversary and Repair Client Messages — Twelve templates for repair-ready and anniversary messages, plus the client segments, consent record and opt-out wording that keep them sendable.
- Using AI Services for Customer Support: Courier Parcel-Query Replies — Reply templates for each tracking status, the triggers that pass a message to a person, and a worked example for a six-driver courier.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: RepairDesk, RepairShopr, Goodcall and Tidio pages, fetched October 2026; Smith.ai and Intercom prices from the pipeline fact sheet, September 2026.