Stay interviews in workers' own language at plants, with a tracked fix list
TL;DR Tenurely sends trained interviewers to factories, warehouses and stores to talk with new frontline workers in their own language before they quit. It turns what they hear into a short fix list (late pay, a harsh supervisor, a broken bus route) and tracks each fix with site HR. Employers pay per site per month.
Frontline jobs in India churn fast. Hunar.AI's 2025 report put attrition in banking frontline roles at 103%, and found entry-level staff would switch for as little as ₹1,000 more a month[1]. A retail HR head estimated retail attrition at 70–80% against a global 30–40%, and said most people leave in their first three months[2].
Every early exit means another round of hiring, induction and lost output on the line. Yet most plants learn why people left only from a one-line exit form, if they learn at all.
Workers rarely tell a supervisor that the supervisor is the problem. Would you? And a lot of them aren't on email or an HR app anyway.
The seed pitched AI that predicts who will quit and steps in. I kept the goal and changed the method, because prediction tools already exist and a score doesn't fix a late salary or a broken canteen. What changes outcomes is finding the site-level cause and fixing it fast.
Tenurely's interviewers, often former shop-floor staff hired from the same region, hold ten-minute stay interviews with workers at day 15, day 45 and day 90. They use a fixed question set in Hindi, Tamil, Odia or whatever the floor speaks, away from supervisors. Answers are tagged into a small set of causes and rolled up by shift and supervisor, never by name.
Each month the plant HR head gets a one-page fix list with owners and dates, and Tenurely checks back on whether each fix actually happened. The proof is simple - did 90-day retention rise in the cohorts after the fix?
The worker is interviewed but the plant or warehouse operator pays. Buyers are HR heads at manufacturers, third-party logistics firms and retail chains with 200–2,000 frontline staff per site. No public price exists for this kind of service, so the sprint must test one. Price it against the cost of a single re-hire, which the buyer can calculate.
The four labour codes took effect on 21 November 2025, bringing mandatory appointment letters, timely wages and double overtime for many frontline workers[3]. A plant that falls short on these now has a legal problem as well as a retention one. The EPFO wage ceiling also rose to ₹25,000 a month from September 2026, changing take-home pay for many workers[4].
There's no public figure for spending on frontline retention services in India. The signal is the size of the churn - 103% frontline attrition in banking roles[1] and an estimated 70–80% in retail[2], with most exits early. Treat the retail figure with care, as it is an older industry estimate.
One Indian startup uses voice AI to hire, onboard, train and engage frontline workers[5]. An employee-listening bot runs AI check-ins for enterprise staff and claims 330+ CHRO clients[6]. Both reach workers through phones and bots. The gap is an in-person conversation on the floor, plus someone who keeps following up until the fix is done.
| Difficulty | Moderate (34/100) |
| Starting capital | ₹40–75 L |
| Time to launch | 3–6 months |
Almost nothing here is technical. The hard part is people - hiring and training interviewers that workers trust, keeping their questions consistent, and getting site managers to act on findings that often point straight at them. Scaling across industrial clusters means running a field team in several states, which is a headache of its own.
No sector licence is needed. Most of the legal weight sits on your client, but you need to know it.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI predict which named worker will quit or decide anything about a person. Reports stay at group level, and people fix the causes.
Want to work out the AI layer for Tenurely? Talk to Yash →
Here's what can go south -
I'd call this a solid service business for a founder with plant HR or field-operations experience. If you want a pure software company then it's the wrong pick. Prove first, at one site, that 90-day retention in the cohort after the first fixes beats the cohort before by at least ten points.
Plan for roughly ₹40–75 L, mostly on hiring and training field interviewers and running them across sites.
No sector licence. You need DPDP-compliant handling of worker data, plus the usual company and GST registration.
It can be, if you price against the cost of re-hiring and clients act on the fix list. There's no public price for this service yet, so test it first.
The free Tenurely validation kit has the line-by-line capital breakdown, the effort scoring, the labour-code and data checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for plant HR heads, and 5 assumptions to test. Download the kit →
Thinking about building Tenurely or something close to it? The make-or-break work is getting site managers to fix what workers tell you, quickly enough to show up in retention. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.