Case study · Healthcare

Members ask and get an answer instantly, on any channel

No queues. No opening hours. No answers that change depending on who picks up. An intelligent agent that serves members with expert judgment, 24/7.

Governed generative AI (RAG)

Single Customer View

True omnichannel

Where it all started

It called itself omnichannel, but it was uncoordinated multichannel

A health mutual serves thousands of members and companies every day: benefits, sick leave, coverage, procedures. Every query, one clear expectation: a fast and correct answer.

The reality was different. Service was split across phone, email, web and front desk, each channel on its own. The same question got different answers depending on who took it. And the same doubts came up a thousand times over, draining the team.

There was no single view of the member and no single source of answers. Multichannel, yes. Omnichannel, no.

The challenge

Repetitive queries that overloaded the team and stole time from what truly needs a person.

Answers that changed by channel and by whoever attended: zero guarantee of consistency.

Member information scattered across several systems, impossible to look up quickly.

Service tied to office hours, with waits right at the volume peaks.

No unified record of what members asked or how it was resolved.

What we proposed

A conversational intelligent agent, built with governed generative AI (RAG) on top of the mutual's knowledge base. It serves members across every channel with a single standard, answers frequent queries instantly, guides each procedure step by step, and hands off to a person at the right moment. Uncoordinated multichannel out; true omnichannel in.

The shift

From the waiting queue to an answer in seconds

The repetitive and predictable moves to the agent. Judgment, empathy and sensitive cases stay where they belong: with people, who finally have the time to handle them well.

Before

Every frequent query resolved by hand, one by one

Different answers depending on channel and person

Service limited to office hours

No single view of members' questions

Now

The agent resolves frequent queries instantly, on any channel

A single source of answers, consistent across every channel

Service available 24/7, with no waits

Every interaction recorded and ready to improve the service

Conversational AI with governed RAG

Judgment, the same answer across every channel

1 %

Immediate answer to the frequent

0 %

The value generated

What changes when service never sleeps

An answer in seconds

Members get what they need on the spot, with no waits and no transfers between departments.

The same expert judgment, always

The quality of the answer no longer depends on where they ask or who is on shift.

The team, freed up

Without the weight of the repetitive, people focus on complex cases and close, personal service.

24/7 service

Basic service stops depending on the clock: members do not wait until Monday.

Every query, intelligence

Questions are recorded and feed the continuous improvement of the service.

On top of what you already have

It builds on the existing knowledge base and current channels, without redoing anything from scratch.

What we achieved

What used to be an endless queue of repeated questions is now resolved in seconds, with the same expert judgment at any hour and on any channel. The team wins back time for what really matters: people. The mutual serves better, faster, and without losing the close, personal touch that defines it.

Member Services Management

Health mutual / Healthcare sector

F&Q

Frequently asked questions

What is an omnichannel intelligent agent?

It is a conversational AI that handles queries across every channel (web, phone, email, messaging) with a single answer standard and a unified view of the member. At BertIA, an AI and data consultancy based in Barcelona, we build it with governed generative AI for healthcare settings, where every answer must be fast, consistent and traceable.

Does it replace the service team?

On the contrary, it strengthens it. The agent resolves the frequent and predictable, and hands off to a person as soon as the case calls for it. The team is freed from the repetitive to focus on complex cases and human contact.

Where do the answers come from? Are they reliable?

It answers exclusively from the mutual’s own knowledge base, not from generic information on the internet. The technique (RAG) ensures every answer comes from the entity’s official sources, with the same consistency the best member of the team would give.

How is members' information protected?

The agent operates within the entity’s security and data protection framework, with access control and traceability of every interaction: a non-negotiable requirement in healthcare.

Does it integrate with what we already have?

Yes. It builds on the existing knowledge base and current service channels, without duplicating systems. It connects to those sources, defines when it hands off to a person, and validates every answer before it goes live to members.