Repair management
Repairs that run themselves
Pango diagnoses the problem, checks the warranty, picks the best outcome and books the right service center. Your team handles only the exceptions.
Book a demoKnow what is wrong before anything ships
From the customer’s own words, photos and audio
A repair case being diagnosed. The customer writes that the left side of their headphones crackles when they turn their head, and attaches a six-second audio clip. The diagnosis agent listens to it and finds a broken cable at the left hinge, 94% sure, matching 312 past repairs, and the product is in warranty. Example data.
The right service center, the first time
Routed by distance, skills, queue and parts on hand.
One repair being routed. The customer is in York. The nearest place, a store in Harrogate nine miles away, has no audio technician. The Leeds workshop, 24 miles away, has the part in stock and a two-day queue. Manchester has a nine-day queue. The repair is sent to Leeds, and a UPS pickup is booked for tomorrow. Example data.
Technicians fix products. Pango does the admin.
Every service center gets its own queue and a co-pilot. Say what you found, and the fault code, parts and notes are filled in.
A technician’s workspace at the Leeds workshop. Its own queue of repairs is on the left; one case is open. The technician says aloud what they found: the left hinge cable was broken, the cable and ear cushion were replaced, and it sounds right. The co-pilot fills in the fault code, the parts used and the tests passed, prints the UPS return label, and the case moves to ready to ship. Example data.
Change policy by saying it
New rule, new market, new product: one sentence, written as a rule, tested on past cases and live in minutes.
Write your first policy in a demoThe policy screen of the Pango app. A policy is said in plain words: headphones are send-in only, never offer drop-off on islands. Pango writes it as two rules, each in plain words and in its formal form, under the rules already there. A back-test replays the last 90 days of cases: 212 cases would change, drop-offs offered on islands go from 31 to none, and repairs sent to the wrong place from 48 to 9. The policy is then published and logged in the audit trail. Example data.
Customers
“A ton of value, and an amazing team that goes above and beyond to solve anything and everything.”
Hakar, CEO, Switch Nails
















One repair case and every way it could end. A remote fix is ruled out, because it is a hardware fault. Sending a part is ruled out, because it is not a home repair. A send-in repair costs £18 and takes four days, an advance swap £64, a credit £129. The send-in repair is chosen: the cheapest outcome that meets the policy. Example data.
Smart resolution
The right outcome, every time
Repair, swap, loaner, part or credit: Pango picks the cheapest one that keeps the customer happy.
- Remote fixes first
- Advance swap
- Loaner during repair
- Trade-in
A retailer’s claim file. A spreadsheet of 248 lines is uploaded and read in four seconds. Each line gets its answer: approved, return requested, or rejected because the serial number was not found. 241 of the 248 lines are approved, and a credit note for £6,214 is issued in the ERP. Example data.
B2B claims
Retailer credits in seconds, not weeks
Upload any claim file. Pango checks every line and closes the loop with your ERP.
- Excel, CSV, email or photo
- Errors flagged per line
- Credit note from your ERP
- Retailer scorecards
Hinge cable failures on one headphone model, by production batch. Four batches fail at about half a percent. The newest batch, B-2207, climbs from 0.7% to 2.4% in 30 days, past the 2% limit. The product team is alerted and sales of the batch are paused. Example data.
Quality intelligence
Every repair improves the product
See which batches fail, why, and what it costs, as it happens.
- By batch, factory, firmware
- Early warnings
- No fault found
- Ask in chat
Built to fit your stack
Pango plugs into your shop, ERP, warehouse and 100+ carriers. Nothing changes on your side.
100+ carriersbooked per parcel
- 100+ carriers
- booked per parcel
- Your stack
- Shop systems · ERP · WMS · REST API · MCP
- ISO 27001 and SOC 2 Type II in progress
- GDPR compliant
Your team keeps its tools. Pango works inside them.
Which part of your operation should work better?






























Book a 15‑minute
demo with a specialist
See how AI would transform your operations.
Questions, answered
Something else on your mind? Talk to a specialist
A broken product from the first complaint to fixed, replaced or recycled. Agents diagnose the problem, check the warranty, pick the outcome, book the right service centre and keep everyone informed, and your team handles the exceptions.
Customers describe the problem in their own words, with photos, video or a voice note, and guided troubleshooting closes the case when a fix works. Pango maps the symptoms to fault codes using your own repair history and gives a confidence score; low-confidence cases go to a person.
By a live lookup against your serial and warranty database, with statutory, extended and goodwill rules for each market, channel and product. Pango reads receipts and asks for a better one when it needs to.
A remote fix, a part sent to the customer, a drop-off or send-in repair, an advance swap, a loaner during the repair, a paid repair, a credit without a return, or a trade-in toward a new product. Pango picks the cheapest one that satisfies the law, your policy and the customer.
Each centre gets its own workspace and sees only its own queue. A case shows the customer’s description, the media and the diagnosis, and technicians log fault codes, parts and tests and print the return label from the bench.
A retailer uploads an Excel or CSV file, a forwarded email or a photo of a list, and it is read in seconds with errors flagged row by row. You approve, reject or request a return for each line, and the credit issued in your ERP closes the case.
By saying it: “Headphones are send-in only. Never offer drop-off on islands.” Pango shows the rule in plain and formal form, tests it on past cases and publishes it once you approve.
You set it for each agent, market and product: suggest, act with approval, act and report, or fully autonomous. Money limits, legal rules and never-do lists are guardrails, and a new rule can run silently on live cases until you trust it.





