01Ordered call lists
Ranked by who will pay, on the channel they answer.
- Propensity and best channel per account, every morning
- Guard-rails on contact hours and scripts built in
Higher early-bucket cure
CommercialAutoCollect · collections agents for lenders
The right borrower, the right channel, the right hour.
For lenders’ collections teams. Accounts ranked by propensity to pay; reminders and payment links on WhatsApp; voice agents that take promises in the borrower’s language; field visits routed for the rest — inside contact-hour rules and approved scripts, with every call summarised and scored.
- +20–35% early-bucket target
- Approved scripts only
- Every call scored
WhySound familiar?
Three things we hear in every first conversation.
From collections heads and managers at banks and NBFCs. Composite quotes from discovery calls — the problems are real; the names are left out.
-
“We call in order of amount, not of who will pay.”
The easiest cures wait behind accounts that will not pay this week, and the bucket rolls forward.
Collections headBucket roll-forward
-
“Most borrowers just need a reminder and a link.”
Agents spend their day dialling people who would have paid from WhatsApp.
Collections managerAgents on the wrong calls
-
“We only hear about bad calls from complaints.”
A fraction of calls are sampled, so script and contact-hour breaches surface too late.
Compliance officerConduct risk
HowWhat changes with AutoCollect
The same three problems, on the screens that solve them.
Product screens with illustrative data. In a working session we design a test against your control group.
02Calls for a reminder
A reminder, a link and a promise — in one chat.
- Payment links and promise dates on WhatsApp
- Broken promises re-queued automatically
Agents free for the hard cases
03Sampled calls
Every call summarised and scored.
- Script, compliance and outcome on every call, people and voice agents alike
- Disputes and hardship always go to a person
Conduct proven, not hoped for
FieldAutoCollect in the field
The people, the phone, the place.
The workplace it runs in, the person who uses it on a phone, the team at the desk and the detail it reads.
Photographs are illustrative — real customer sites are not shown.
- +20–35%Early-bucket curetarget vs ordered call lists
- 6–8 weeksTo first valueon one portfolio
- 8 am – 7 pmContact hoursenforced on every channel
- Every callSummarisedand scored for compliance
- Hindi · English · moreLanguagesborrower’s choice
- HardshipTo a persondisputes never automated
01The screens
From a call list to promises that are kept.
The ranked worklist, the borrower’s WhatsApp and the scored calls.
autowhat · AutoCollect / Collections today
Illustrative data
Accounts ranked by propensity to pay, with the channel and next action — inside guard-rails.
Product screens with illustrative data — every name, place and figure is a placeholder.
02Before and after
Call lists roll accounts into the next bucket.
Ordered call lists treat every borrower the same; the bucket rolls forward while agents dial.
| Area | Today | With AutoCollect |
|---|---|---|
| Worklist | Ordered by overdue amount | Ranked by propensity to pay |
| Channel | Calls only | WhatsApp, voice agent, person or field — by borrower |
| Promises | In notes | Recorded, reminded and tracked |
| Compliance | Sampled | Every call scored; contact hours enforced |
| Hardship | Missed | Detected and sent to a person |
03Ranked, reminded, recorded
Eight steps, from due date to cure.
Your policy decides what can be offered; AutoCollect makes sure the right borrower hears it at the right time.
-
01
Load
LMS
Due and overdue accounts loaded daily from your LMS.
-
02
Rank
Propensity
Accounts ranked by likelihood to pay and the best channel.
-
03
Remind
WhatsApp
Reminders and payment links before and after the due date.
-
04
Call
Voice agents
Promises taken in the borrower’s language, within contact hours.
-
05
Escalate
People
Disputes, hardship and restructuring go to a trained person.
-
06
Visit
Field
Late buckets routed to field agents with proof of visit.
-
07
Reconcile
Payments
Payments matched to promises; broken promises re-queued.
-
08
Score
Compliance
Every call summarised and scored against approved scripts.
04The platform
Worklist, channels, field and compliance.
Start with one portfolio and bucket; the rest join on the same engine.
Worklist · 4
- Propensity ranking
- Bucket strategy
- Promise tracking
- Re-queue rules
Channels · 4
- WhatsApp reminders
- Payment links
- Voice agents
- Agent desk
Field · 4
- Visit routing
- Proof of visit
- Receipts
- Cash deposit match
Compliance · 4
- Contact-hour rules
- Approved scripts
- Call scoring
- Do-not-call list
05How we deliver
One bucket, one metric, six to eight weeks.
On the LMS and dialler you already run.
- Weeks 1–2 · MapPortfolio, buckets, policy and scripts mapped; the cure-rate baseline taken.
- Weeks 2–6 · BuildRanking, WhatsApp, voice agent and compliance scoring on your stack.
- Weeks 6–8 · Go liveOne bucket live, measured against a control group.
- Then · ExtendLater buckets, field and other portfolios.
FAQQuestions we get about AutoCollect
Straight answers, before the call.
AutoCollect is an Autowhat AI product for collections heads and managers at banks and NBFCs. Short, factual answers for buyers, procurement and the assistants people ask first.
In short. AutoCollect is an Autowhat AI product for collections heads and managers at banks and NBFCs. It is built for collections heads and managers at banks and NBFCs. It exists to remove the problems those teams describe in almost every first conversation: call lists in order of amount, not of who will pay; agents dialling borrowers who only needed a reminder; and conduct issues found through complaints. With AutoCollect, teams get accounts ranked by propensity and best channel every morning, whatsApp links and promises; broken promises re-queued, and every call summarised and scored; hardship to a person. It connects to Your LMS, WhatsApp & payment links, Voice agents and Dialler, is used by Collections · agents · field, and runs on whatsapp + voice + field. Like every Autowhat AI product, AutoCollect goes live on one site, lane or process first, usually within weeks, on managed cloud, in your own VPC or on-premises, and shares one operational record with the other products, so the second deployment reuses the data and integrations of the first.
Who is AutoCollect for?
Collections heads and managers at banks and NBFCs. Early-bucket collections ranked by propensity to pay; reminders and payment links on WhatsApp; voice agents that take promises in the borrower's language; field visits for the rest — inside contact-hour rules and approved scripts, every call scored.
What problems does AutoCollect remove?
call lists in order of amount, not of who will pay; agents dialling borrowers who only needed a reminder; conduct issues found through complaints.
What do we get with AutoCollect?
accounts ranked by propensity and best channel every morning; whatsApp links and promises; broken promises re-queued; every call summarised and scored; hardship to a person.
What does AutoCollect connect to, and who uses it?
It connects to Your LMS, WhatsApp & payment links, Voice agents, Dialler. Users: Collections · agents · field. Channel: WhatsApp + voice + field.
How do we start with AutoCollect?
With a 30-minute working session on your own data — an engineer from the relevant sector, not a sales queue. The first wave is scoped in that call; most products go live on one site, lane or process within weeks and then scale.
Where does AutoCollect run, and who owns the data?
On managed cloud, in your own VPC (AWS, Azure or GCP) or on-premises, with your keys. SOC 2 Type II and ISO 27001 controls, India data residency when required, and local or open-weight models when policy demands it. The data stays yours.
06Working session
Bring one month of bucket-1 data. We’ll show you who should have been called first.
Thirty minutes: your buckets, scripts and contact rules — and a test design against a control group.