Bill reminder voice bots for utilities, in every Indian language

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

The scale problem utilities face

An electricity, water or gas utility serving a large customer base across a state has millions of bills due every cycle. A human calling team, even a large one, cannot realistically call every overdue or upcoming-due customer, AI voice calling can run that volume consistently, at the moment it's needed each billing cycle.

Because billing cycles repeat every month on a predictable schedule, this isn't a one-off campaign, it's an ongoing operation that needs to run the same way, reliably, every cycle without re-briefing a calling team each time. That consistency is exactly what a configured voice flow gives you once it's set up correctly.

Many customers currently receive only an SMS reminder, which is easy to miss among promotional messages, or no reminder at all until the bill is already overdue. A voice call is harder to ignore than a text and gets through even to customers who don't actively check their SMS inbox.

Delayed payments also create real downstream cost for a utility, cut-off and reconnection processes, follow-up field visits, and revenue recognised later than it should be. A reminder that actually gets heard, in a language the customer is comfortable in, reduces how often these downstream processes get triggered at all.

Language choice comes first

The call opens by offering a short language menu, press 1 for Hindi, 2 for Marathi, 3 for English, and so on, using a DTMF menu voice block. Getting this right at the very start matters enormously for a utility customer base that spans a huge range of language comfort, especially in semi-urban and rural areas.

Order the language options by how common they are in your service area, not alphabetically, a customer base predominantly speaking one regional language should have that language as option 1, not buried third or fourth in the menu where a hesitant customer might hang up before hearing it.

Stating the bill clearly

Once the language is set, the AI states the amount due and the due date, pulled as variables from your billing system so every call reflects the actual account, not a generic reminder. This is followed by a short explanation of how to pay, via the WhatsApp link that follows, UPI, or at a collection centre.

Keep the script to one clear idea at a time rather than a long paragraph read aloud, state the amount, pause, state the due date, pause, then explain the payment options. A voice call that reads like a wall of text is harder to follow than the same information delivered as short, sequential statements, especially for older customers less used to automated calls.

DTMF confirmation and payment link

A DTMF menu lets the customer press 1 to receive the payment link on WhatsApp immediately, or 2 to speak to a billing officer for a dispute or query. This keeps the call itself short while still giving a clear next action.

Add a third option for customers who've already paid, "press 3 if you've already paid this bill", since a reminder call reaching someone who paid a day or two ago and the system hasn't yet reconciled is a common, avoidable source of frustration if there's no easy way to say so on the call itself.

Building it on Convo360

1. Connect your telephony provider, Exotel, Plivo, Twilio or your own SIP trunk, under Voice settings.
2. Build a voice flow starting with a DTMF menu for language selection.
3. Use Set Variable to pull amount and due date from your billing system via HTTP/API request, then state them in the selected language.
4. Add a second DTMF menu for "send payment link" or "speak to officer", the latter using the transfer block.
5. Trigger calls in batches via the API from your billing cycle system.

See voice docs and compliance docs.

Run a small pilot batch, a few thousand calls in one language and one billing segment, before scaling to the full customer base for a cycle. This surfaces script or language-quality issues while the fix is cheap, rather than after a full-scale run has already gone out with the same flaw repeated across every call.

Handling high call volumes and what to review

At utility scale, calls need to go out in batches spread across the day rather than all at once, both to respect calling-hour rules and to avoid overwhelming your telephony provider's concurrent call capacity. Stagger the API trigger by geography or billing segment so the load is predictable rather than a single massive burst at the start of a cycle.

Review the DTMF menu selection data monthly: what share of customers press 1 for the payment link versus 2 for an officer versus simply not responding at all. A high non-response rate on a specific segment (say, a particular language or region) usually points to a script or language-quality issue worth fixing rather than a payment behaviour problem.

What it costs

AI voice calling needs the Pro plan (₹2,999/month, ex-GST) or above plus a connected telephony account. Call minutes are billed by your telephony provider directly at their rate, and Convo360 adds no markup, a meaningful factor at the call volumes utilities typically run.

Example conversation

Frequently asked questions

Related reading

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