Case study · Voice AI · HR shared services
The voice agent that writes the ticket while you talk.
An HR services provider runs HR for more than fifty companies. When an employee at one of them has a problem, they ring the support line. We built the voice co-worker that answers that line at any hour, in Hindi, English or both at once, works out what is actually wrong and has the ticket written before the caller hangs up.
- The problem
- Queries were never written down, so reporting showed who had a quiet afternoon, not real demand.
- What we built
- A voice and messaging co-worker that answers the line and logs every query.
- The result
- Every query logged and routed to the team that owns it.
- Status
- Live since 2026.
The same problem shows up in
- Payroll bureaus
- Managed IT services
- Facilities management
- Member associations
- Franchise support desks
If your service desk only has a record of the queries someone remembered to write down, your reporting is a picture of who had a quiet afternoon, not of demand.
Before and after.
Each week one person carried support duty on top of their own job, and every answered call still had to be typed into a list afterwards.
- 1Answer the phone, if you are at your desk
- 2Understand the problem
- 3Finish the call
- 4Remember to open the list
- 5Type the query in
- 6Work out whose job it is
Four ways a query died: answered and never logged, logged later, a missed call, or an after-hours call nobody heard.
The line is answered on the first ring, at any hour, and the query is written down while the employee is still on the phone.
- 1Ring the same number as always
- 2Say what is wrong
No longer done by hand
- Finish the call
- Remember to open the list
- Type the query in
- Work out whose job it is
The record is a by-product of the conversation instead of a task that comes after it.
- 50+
- Client companies
- 3
- Languages
- 24/7
- Answered
- 2
- Channels
Served from one support line.
Hindi, English and Hinglish, even in one sentence.
First ring, any hour.
Phone and messaging.
The ask was smaller than the problem.
They asked for a way to capture queries. What they needed was for capture to stop being a separate act of remembering.
We mapped their day first
Discovery sessions on how support actually ran, then a story map of every step.
They could see it before they bought it
The map let them recognise their own problem in the plan before any code existed.
We split it in two
Phase one captures and routes every query. Phase two resolves them from each client's policies.
Phase one had to earn phase two
Resolution is only added once logging is trusted.
Capture first, resolve next.
Answers on the first ring
At any hour, in the caller's own language mix.
Works out what is wrong
Asks only what it needs and checks it understood.
Writes the ticket live
The record exists before the caller hangs up.
Routes to the owner
Each query reaches the team that owns it.
Start with one process.
Tell us which part of your work takes the most time. We'll show you what an AI co-worker could take off your team's plate.
Contact us
We'll use your details to reply to your enquiry. See our Privacy Policy.
What to expect
We start with one process and put something real in front of your team within weeks, so you can judge it on your own work.
Thor Signia
How we start
