What an AI agent can and cannot do with your enquiries

An AI agent is software that reads a message from a customer, works out what is being asked, and replies in plain language. Modern agents are built on large language models, and when they are connected to a business’s own information, such as its services, prices, opening hours and policies, they can answer a large share of everyday questions without a person typing the reply.

Where an agent helps

  • Routine questions at any hour. What do you offer, how much does it cost, how do I book, where are you. These make up much of a typical inbox, and they have clear answers.
  • Qualifying an enquiry. An agent can ask the questions a salesperson would ask first, such as budget, timing and what the customer needs, so that the person who follows up starts with the full picture.
  • One answer across channels. The same agent can work in a website chat, by email and in messaging apps such as WhatsApp, so a customer gets a consistent answer wherever they write.

Where it should not decide alone

  • Anything it has not been told. An agent should answer from the business’s own information. When the answer is not there, the right behaviour is to say so and pass the conversation to a person, not to guess.
  • Commitments. Discounts, refunds, contract terms and anything with legal or financial weight belong with a person.
  • Complaints and sensitive situations. A frustrated customer wants to be heard by someone who can act. A good set-up recognises these conversations and hands them over quickly.

What to set up first

  1. Write down the answers. The agent is only as good as the information it is given. A short, accurate document about services, prices and policies is worth more than any setting.
  2. Decide the handover. Who receives the conversations the agent cannot handle, on which channel, and how quickly they reply.
  3. Tell customers it is an AI. People accept an assistant that is honest about what it is, and it sets the right expectations.
  4. Read the conversations. In the first weeks, review what customers asked and where the agent fell short, and add the missing answers.

Measuring it

The useful measures are simple: how many enquiries were answered without a person, how many were passed on, how quickly people received a first reply, and how many enquiries became customers. Those figures, tracked month by month, show whether the agent is doing its job.

AI agents for enquiries are one part of our King-OP platform. If you are considering one for your business, contact us.

More insights

Tell us what you are working on

Send a short description and we will reply by email with the next step.

Contact us