A loan enquiry chatbot for WhatsApp: eligibility, documents, callback

Updated 2026-09-19 · 4 min read · By the Convo360 team

Why loan enquiries need a fast first response

A prospective borrower messaging your NBFC's WhatsApp number about a personal or vehicle loan is usually comparing two or three lenders at the same time. If your reply takes hours, they've likely already spoken to someone else by the time it arrives.

A chatbot answers the first round of questions, product type, rough eligibility, documents, instantly, keeping the enquiry warm until an officer can call.

The enquiries themselves are rarely complicated at this stage. Most borrowers just want to know if they're broadly in range before sharing more details. A bot that answers this honestly, without over-promising, actually builds more trust than a generic "someone will call you soon" auto-reply that doesn't move the conversation forward.

Many enquiries also arrive well outside office hours, in the evenings and on weekends, when someone finally has time to sit down and think through a purchase or an expense that needs financing. An NBFC only staffed for daytime calls loses this entire segment to whichever lender happens to answer first the next morning.

The product menu and eligibility questions

A List Menu block presents the loan types you offer. Based on the selection, the flow asks 3-4 qualifying questions: approximate monthly income, employment type, loan amount needed, and city.

• Personal loan: income, employment type, amount
• Vehicle loan: vehicle type, on-road price, down payment planned
• Business loan: business vintage, monthly turnover, amount
• Gold loan: approximate gold weight, amount needed

These answers are enough for a rough eligibility read, not a final decision.

Keep the eligibility statement conditional and clearly worded: "Based on what you've shared, this is broadly within our range for a personal loan. Our officer will confirm exact eligibility after reviewing your documents." Avoid a flat "you're eligible" or "you qualify", since either phrase can be read later as a commitment the firm didn't actually intend to make at the enquiry stage.

Documents, without a human repeating themselves

An AI Reply block grounded in your knowledge base answers document questions specific to the loan type: PAN, Aadhaar, salary slips, bank statements, ITR for self-employed applicants. This alone removes a large share of the "what documents do I need" calls your officers field every day.

Structuring the checklist by employment type inside the knowledge base, rather than one long combined list, keeps answers relevant: a salaried applicant doesn't need to read through GST return requirements meant for the self-employed segment, and vice versa.

What the bot must never say

The flow should never state or imply an approval probability, a guaranteed interest rate, or a specific sanctioned amount, since these are underwriting decisions. Every response about outcomes should point to "our loan officer will confirm this based on your documents", keeping the bot factual and compliant.

A related mistake is letting the AI Reply block improvise on interest rates when asked directly. "What's your interest rate?" should trigger a fixed, pre-approved range from the knowledge base, never a number the model generates on its own, since even a slightly wrong figure quoted in chat can create a real dispute later.

It also helps to set the AI agent's system prompt explicitly: instruct it to never use words like "guaranteed", "approved" or "pre-qualified" in any reply, and to always redirect specific eligibility questions to the human callback. This is a one-time configuration step that removes an entire category of compliance risk from every future conversation.

Building it on Convo360

1. Upload your product sheets, eligibility criteria and document checklists as an AI agent knowledge base.
2. In Flow Builder, start with a List Menu for loan type, followed by Ask a Question blocks for income, amount and city.
3. Use Condition blocks to score the lead as hot/warm/cold based on income and amount, storing the score with Set Variable.
4. Save every qualified lead with the Save CRM lead block into your CRM pipeline, tagged by loan type and score.
5. End with "Our loan officer will call you within 24 hours" rather than any approval statement.

See AI agents docs and CRM docs.

Prioritising callbacks with lead scoring

Not every enquiry deserves the same urgency. On Convo360, a Condition block can score each lead hot, warm or cold based on income band and loan amount against your typical approval range. A ₹15 lakh vehicle loan enquiry from someone with clearly insufficient stated income is still worth a courteous callback, but shouldn't sit at the top of an officer's queue ahead of a well-matched applicant.

Review the CRM pipeline weekly for one pattern in particular: enquiries that stall at the document-checklist stage without ever reaching a callback. That usually means the checklist itself is discouraging applicants (asking for more than is really needed at enquiry stage) rather than a lead-quality problem.

What it costs

This runs comfortably on the Starter plan (₹999/month, ex-GST) for the flow, knowledge base and CRM; adding it to voice as well, so officers can call leads back through the same platform, needs the Pro plan (₹2,999/month, ex-GST). Voice calls are billed by your own telephony provider at their published rate, with no per-minute markup from Convo360, so the extra cost stays predictable as call volume grows.

Example conversation

Frequently asked questions

Related reading

Available across India — see WhatsApp & AI calling automation by state and city.