A shoe repair shop needs software that gives every job a unique ticket number, records the item, work, quote and promised date, texts the customer at quote and at pick-up, and flags what is left on the shelf. AI helps by drafting those texts and policy wording from your templates, never by pricing or promising.
Most shops need less than they think. The software question is really a counter question: what do you write down when a pair of boots arrives, and what do you say when they are ready? Get those two moments right, and almost any system, or a spreadsheet, will do.
What goes on a job ticket
A ticket is the shop's memory. If it is thin, you will phone customers to ask what they dropped off. If it is thorough, a new member of staff can finish the job without asking anyone. This is the field list I would build the system around.
TICKET 26-0412
Customer: [first name], mobile ending 0187, OK to text: yes
Item: black leather ankle boots, size 40, left and right, brand not visible
Condition at drop-off: scuff on left toe, sole worn at heel, laces missing
Work requested: full heel replacement, re-glue right sole edge, polish
Quote: $48, approved at counter (yes/no): yes, deposit $15 taken
Promised: Thursday 5 November, after 12:00
Tag number on item: 26-0412 (two tags: one inside, one on the bag)
Photos: 2 taken (both sides, soles)
Staff: taken by [initials], repaired by [initials]
Notes: customer wants original heel taps kept
Status: receivedEach field earns its place. Condition at drop-off protects you if a customer later disputes damage. Photos settle most of those disputes in seconds. The tag number on the item, not only on the paperwork, is what prevents the worst error in this trade: a pair going home with the wrong person. Consent to texts, recorded at the counter, lets you message without awkward questions later.
Ticket numbers that do not clash
Use a prefix plus a running number, for example the two-digit year then four digits: 26-0412. Never reset to 1 each week. Print the number on a tag, write it on the item, and keep the same number in every text. If you run two counters, give each its own series (26-A0412 and 26-B0098) so two staff never create the same number at the same moment. A good system generates the number for you. On paper, a numbered ticket book does the same job.
Status stages that match how the bench works
Keep the stage list short. Every extra stage is another click at a busy counter, and staff will skip it.
- Received. Item tagged, ticket made, photos taken.
- Quote sent. Used when the price is not agreed at the counter.
- Approved. Customer said yes; work can start.
- In progress. On the bench.
- Waiting on a part. A heel, fastener or sole is on order.
- Ready. On the shelf, text sent.
- Collected. Paid and handed over.
The two stages that matter most are Quote sent and Ready, because both are moments when the customer must act. Everything else is for the team.
The quote text, and where AI helps
Some jobs cannot be priced at the counter: a boot that needs a new fastener, a handbag with a broken strap, a sole that may not survive the glue. You inspect it later and text a price. A quote text should say what you will do, the price, the date, and how to say yes. It should never say "approx" without a range.
Hi [first name], it's [shop name]. Quote for ticket 26-0412 (black ankle
boots): full heel replacement and re-glue of the right sole is $48,
ready by Thursday 5 November. Reply YES to go ahead or call us on
[number] to chat. No reply means we won't start the work. Thanks.Notice the last sentence. A quote that says what happens if the customer stays silent saves arguments. If you hold a shoe for 14 days without an answer, the next text should say so.
A prompt that drafts the text from the ticket
This is where AI earns its keep. Paste the ticket fields, then use a fixed instruction. Never let it set the price.
You write short text messages for a shoe repair shop. Use only the facts below.
Do not invent prices, dates or work. Keep it under 50 words, polite and plain.
Include the ticket number, the item, the price I give you, the ready date, and
how to say yes. If a fact is missing, write [MISSING: fact] instead.
Ticket: 26-0412. Item: black ankle boots. Work: heel replacement, re-glue
right sole. Price: $48. Ready: Thursday 5 November. Customer first name: [first name].Illustrative output: "Hi [first name], quote for ticket 26-0412, black ankle boots: heel replacement and right sole re-glue is $48, ready Thursday 5 November. Reply YES to go ahead, or call us to discuss." What you would fix: add the shop name, because a text from an unknown number may be ignored, and add the line about silence. If the model wrote "approximately $48" or "possibly Friday", delete both, since you gave it exact facts. Give the assistant a standing description of your shop so the tone fits; the context tutorial shows how.
The pick-up text, and the reminders that follow
When the job is ready, send one clear text: ticket number, item, amount to pay, opening hours, and what to bring. Then set up a reminder rhythm for items that sit on the shelf.
| When | Message purpose | Example wording |
|---|---|---|
| Day 0 (ready) | Tell them it is ready | "Your boots (26-0412) are ready. $33 to pay. Open 9 to 5:30 Tuesday to Saturday." |
| Day 7 | Friendly nudge | "A reminder that 26-0412 is waiting for you. We're open until 5:30 today." |
| Day 30 | Mention the storage policy | "26-0412 has been ready for 30 days. Please collect by [date] as our policy says." |
| Day 60 | Final notice before the policy applies | "Final reminder for 26-0412. If we hear nothing by [date] we will follow the unclaimed-item notice on your ticket." |
Adjust the days to your shelf space. A small shop might use 7, 21 and 45. The key is that each text matches what the ticket said at drop-off, so the final notice is no surprise.
Unclaimed-item policy wording
Every shoe repairer ends up with a shelf of orphaned pairs. The solution is to put a notice on the ticket and the counter, and then follow it. The wording below is a starting point for your own policy. Rules on abandoned property differ from place to place, so have a qualified adviser check what you may do after the notice period before you use it.
Uncollected items: We will text you when your item is ready. Please collect
it within [60] days of that text. After that period, and after we have tried
to contact you at the number on this ticket, we may [donate, recycle or sell]
the item to cover our costs. By signing this ticket you agree to this notice.
Storage charges: none for the first [30] days, then $[amount] a week.Square brackets are deliberate: the period, the action and any storage charge are your decisions and your adviser's. Ask AI to tidy the grammar and to produce a shorter version for a counter sign, but not to decide what the law allows. A useful prompt is: "Rewrite this notice at a reading age of 12, keep every number and condition exactly, and do not add promises."
Jobs that stall: parts, rush orders and deposits
Waiting on a part
A job that needs a special sole or a matching fastener stalls, and silence is what annoys customers. Move it to Waiting on a part, add the expected arrival date to the ticket, and send one text at the moment it stalls rather than waiting to be asked. Illustrative wording: "Hi [first name], ticket 26-0412 needs a replacement sole that is on order. We expect it by Tuesday 10 November and will text you the day it's fixed. If you need the boots sooner, call us and we can return them as they are." The last sentence gives the customer a choice, which is often what turns irritation into patience. Check how your software handles a date change: if the part arrives late, a second text should go out automatically or be on the next morning's list.
Rush jobs
Many shops charge extra for same-day work. Put the rush fee on the ticket as its own line so it shows on the receipt, and add a "rush" flag that moves the ticket to the top of the bench list. Decide your rush rule in advance: for example, rush is only offered before 11:00, and only for work on your fast-jobs list (heels, simple stitching, polish). A rule written down stops the counter from agreeing to jobs the bench cannot do.
Deposits on expensive repairs
A resole on boots costing several hundred dollars is a different risk from a heel tap. A common practice is to take a deposit on larger quotes, and record it on the ticket with the balance due. Say plainly on the ticket whether the deposit is refundable if the customer cancels after work has started, and ask a qualified adviser to review that line, since refund rules vary. If your system tracks balance due, the pick-up text can state the exact amount, which prevents a surprise at the till.
Asking for a review without being pushy
The best moment to ask for a review is a few hours after collection, when the shoes are still a pleasure. A text that follows the same template each time is enough: "Thanks for collecting ticket 26-0412, [first name]. If you were happy with the repair, a short review helps other people find us: [link]." Have AI draft three variations so that your messages do not sound identical, and keep each under 30 words. Only ask customers who collected and paid happily. Never send one to a customer who raised a problem. The garage status updates tutorial and the phone repair support tutorial show how other repair trades handle the same rhythm of quote, ready and follow-up.
Handling a complaint about finished work
Every shop gets the call that says a heel has come away or the colour is wrong. The ticket is what lets you answer calmly. Pull up the number, read the work requested, look at the drop-off photos, and decide whether the issue is workmanship (fix it free) or wear and tear on old leather (explain, and offer a quote). Draft the reply with AI if it helps you stay polite, but give it the facts and the decision.
Customer message: "The heel on my boots came off after two weeks."
Facts: ticket 26-0412, heel replaced 12 days ago, photos show original
glue line intact on the right boot, left boot heel base cracked before work.
Decision: re-fit the left heel free of charge; ask customer to bring both pairs in.
Draft a reply under 70 words, apologetic but not admitting more than the facts.Illustrative output: "Sorry to hear that, [first name]. Please bring both boots in with ticket 26-0412 and we'll look at the left heel straight away; we'll re-fit it at no charge. We're open 9 to 5:30 Tuesday to Saturday." What you would fix: remove any line where the model apologises for something you do not accept, and confirm the opening hours are right.
A prompt for the weekly shelf review
Once a week, export or copy the list of tickets still on the shelf, with their ready dates, and ask AI to sort them into who needs which reminder. This saves the manual hunt and makes it easy to hand the list to a part-time member of staff.
Here is a list of repair tickets with ticket number, item, status and the
date the item became ready. Today is [date]. For each ready item, tell me
how many days it has waited and which reminder is due: none, day 7, day 30 or
day 60. Output a table, then draft each message using only the ticket facts.
Do not change any dates or amounts.Illustrative output: a table with ten rows, of which three need the day-7 nudge and one is at day 60. What you would fix: check the day count yourself for the two oldest items, since models occasionally miscount dates. Keep the final say with a person.
Does a shoe repair shop need dedicated software?
Here is a simple scale. Count your jobs a week and your pain points, and pick the lightest tool that solves them.
| Shop size | Enough | Add when |
|---|---|---|
| Under 25 jobs a week, one counter | Numbered ticket book, tags, a phone and a spreadsheet | You miss texts or lose a ticket twice in a month |
| 25 to 80 jobs a week | A repair-shop system with ticket numbers, status and text messages | You want photos on the ticket and a shelf report |
| 80+ jobs a week, several staff or sites | A system with user logins, barcode tags and reports by branch | You need audit trails, timekeeping or accounting links |
When you look at products, you will meet two kinds. A few are built for counter trades that tag items: Arbelsoft, for example, sells ShoeMax alongside its dry-cleaning and tailoring systems, with drop-off and pick-up screens, tickets and tags, caller ID and QR codes, sold as an installed package with hardware. Many others, such as RepairTree or RepairDesk, were built for phone and electronics repair; they cover ticketing, ready-for-pickup texts and online booking well, but their device fields (model, IMEI, fault) need renaming for heels and soles. Neither kind publishes a price that suits every shop, so check each vendor's own pages for pricing, text-message charges and whether the ticket can carry photos and custom fields such as heel type or leather colour. Then run a trial with ten real tickets, not a demo.
A buying checklist
- Can I print a tag or sticker with the ticket number and a barcode?
- Can customers be texted from the ticket, and is the sender my shop's name or number?
- Can I add my own fields (shoe size, leather colour, "keep old heels")?
- Does it store photos against the ticket?
- Can I see everything on the shelf by age?
- Is there a record of who changed a status and when?
- What are the fees per text, and what happens to my data if I leave?
The last question is the most overlooked. Make sure you can export your customer and ticket history as a file, because your history of regulars is valuable.
A worked example: a two-person shop with 40 jobs a week
An illustrative shop has two staff and around 40 jobs a week. About 28 are priced at the counter, 12 need a quote text. Each quote text takes about 3 minutes to write by hand, so 12 x 3 = 36 minutes. Pick-up texts take around 2 minutes each, so 40 x 2 = 80 minutes. Reminders for the 8 or so items left on the shelf each week add roughly 24 minutes. In total that is close to 2.4 hours a week of typing.
With templates and an AI drafting step, a quote text takes about 1 minute (paste facts, check, send), a pick-up text about 30 seconds, and reminders are batch-drafted in 5 minutes. The new total is about 12 + 20 + 5 = 37 minutes, a saving of roughly 1.8 hours a week, or about 94 hours a year. If your time is worth $25 an hour, that is around $2,300 a year of counter time. Against a system fee of, say, $50 a month ($600 a year), the maths is comfortable, and the larger benefit is that nobody forgets to text.
These are assumptions, not measurements. Time yourself for a week before buying anything.
Other places AI is useful at the counter
Turning a messy note into a clean ticket line
Staff write notes in a hurry: "L heel gone + sole lift, R ok, cust wants vibram not leather". Paste a few notes into an AI tool and ask for a tidy line for the ticket and for the bench. Illustrative output: "Left heel: replace (rubber requested). Left sole: re-attach lifting edge. Right: no work. Customer prefers a rubber sole over leather." You check that "vibram" is a brand rubber and keep what the customer asked for.
Answering the repeated questions
Customers ask the same things: "how long for a heel?", "can you stretch these?", "do you do handbags?" Write the answers once in a short FAQ and have AI turn them into replies for your inbox and website. The estimate follow-up tutorial has wording for chasing a quote that nobody has answered.
A before and after for a difficult text
Before: "Your shoes are ready but there is a problem, call us." After: "Hi [first name], ticket 26-0412 is ready to collect, but the leather on the right sole was thinner than expected, so we've held off the extra glue. No extra charge yet. Please call [number] and we'll explain." The second text says what, why and what happens next, which is what a worried customer wants.
A mistake that happens in real shops
Illustrative: two pairs of brown loafers arrive on the same morning under separate tickets. Only the ticket paperwork carries the numbers, not the shoes. At collection, the wrong pair goes home. The fix is cheap: tag the item itself, take a photo, and say the last four digits of the number aloud at hand-over. Software helps by printing the tag, and AI does not help at all here, which is a useful reminder that the process matters more than the tool.
Three shelf numbers to count after a month
After a month, count three things. How many jobs sit on the shelf more than 30 days (the number should fall). How many customers phoned to ask "is it ready?" (it should fall, because texts reach them first). How many jobs had a dispute about condition or price (it should be near zero, because the ticket and photo settled it). If these numbers do not move, the problem is the habit, not the software: staff are skipping the photo or the text. Start by making those two steps part of closing the ticket, so a ticket cannot be moved to Received without them.
Questions shoe repairers ask about software and AI
Do I need dedicated shoe repair software?
Not always. A shop with a few dozen jobs a week can run on numbered paper tickets plus a spreadsheet and a text-message tool. Move to software when you lose tickets, forget to text customers, or cannot say what is waiting on the shelf without counting it.
Can AI write the pick-up texts?
Yes, as long as it only fills a template you have approved. Give it the ticket number, item and status, and have it draft a short message you check before sending. Never let it invent prices, dates or promises that are not on the ticket.
How long should an unclaimed item be kept?
That is a policy and legal question, because rules on abandoned property differ between places. Write a clear notice on the ticket, send reminders, and ask a qualified adviser what you may do with items after your chosen period before you rely on the wording.
What should a ticket number look like?
Short, unique and printable on a tag: a prefix for the year or month plus a running number, such as 26-0412. Avoid starting again from 1 each week, because duplicate numbers on the shelf cause wrong hand-overs.
Do customers need to consent to texts?
Ask at the counter and record it on the ticket. Tick a box for 'text me when ready', and keep texts to the job in hand. Marketing messages are a separate permission and rules vary, so check with an adviser.
Further reads
- AI for a Shoe Repair Shop: Quotes and Ready-for-Collection Texts — More quote and collection text wording for cobblers.
- Phone Repair Intake Form: Turn Customer Messages Into Tickets With AI — An intake form and ticket flow from a sister trade.
- AI for Dry Cleaners: Damage Claims, Stain Notes and Customer Replies — Condition notes that settle damage claims.
- AI App for Tailors: What Measurement and Fitting Apps Can Do — Another made-to-fit trade, and what apps do for it.
- How Dry Cleaners and Laundries Use AI for Orders and Collections — Collections and uncollected items, solved for laundries.
- Music Store POS System: Serial Numbers, Rentals, Repairs and AI — Serial numbers and repairs in a music shop POS.
- How to Get More Google Reviews With Automated Review Requests — Automate review requests after collection.
- Repair Shop Software for Bike Shops: Service Bookings and Quotes With AI — Service tiers, intake questions, a spring-rush booking cap worked out in bench hours, and a quote prompt that only uses your own price list.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: Arbelsoft product pages (ShoeMax), RepairTree and RepairDesk feature pages, checked October 2026. No prices are quoted; check each vendor's own pricing page before buying.