For businesses

AI customer assistant for telecom and internet providers: usage and plan questions answered from billing

How a carrier configures Kav: data-usage and plan questions answered from the billing system as a real table, with a plan change offered in the same reply.

By The Kav team Published 2 min read

Telecom contacts cluster around three questions: how much have I used, what am I paying, and which plan fits me. Each one is a lookup in a system the carrier already runs, and each one is asked most often when the customer is worried.

A depiction, not a customer. No carrier is named here. The example is the platform's own illustration, marked example data on the home page.

What a subscriber asks, and what Kav answers

  • "How much data have I used this month?" The answer is usage by day, drawn as a real table: a grid a screen reader can move through row by row and column by column. The capability declares that its output is tabular and the widget draws the columns its author named; it never guesses a layout from the numbers (why a table is a table).
  • A plan change at the moment of need. Add a data pack and Compare plans sit beside the usage answer.
  • The same answer everywhere. The example runs on the website; the same capability serves the other channels (which channels does Kav support).
  • The bill question. "Why is my bill higher?" works the same way, from the same billing system (the bill-spike walk-through).

Where the money moves

The bold rows are the levers a carrier leans on most.

Lever What Kav does here What you measure
Fewer repeat contacts Usage, balance and plan questions are answered from billing. Usage and bill contacts as a share of all contacts.
Faster when a person steps in A plan-change request reaches an agent with usage already on the screen. Handling time on plan changes.
Out of hours, every channel Usage worries peak near month-end and at night; the answer is there then. Contacts outside office hours around billing dates.
Fewer errors and rework Usage figures come from billing through a typed template and are never typed by the model. Disputes traced to a mis-stated figure.
Revenue kept or captured A data pack or plan comparison sits next to the usage answer, at the moment the customer needs it. Data-pack and plan-change conversions that start in a chat.

Work out your own number

  1. Separate usage, bill and plan contacts from the rest of your contact log for four weeks.
  2. Cost per contact = agent hours × loaded hourly cost ÷ contacts handled.
  3. Saving = lookups answered without a person × cost per contact. Revenue = data packs and plan changes started in chat × your margin on each. Keep the two apart.

Work out your break-even from the published prices, or talk to us about connecting your billing system.