95% Shorter Onboarding and Daily Use Instead of Quarterly – Embedded AI Customer Portal for a Cybersecurity Platform Serving Fortune 500 Clients (USA)

We redesigned the customer-facing panel of a US cyber risk platform around an embedded AI agent. Onboarding dropped from 45 minutes with support to under 2 minutes self-serve, and a tool customers opened once a quarter became a daily decision-support system. Board members, CFOs and CISOs use the same interface, each in their own way.

The Challenge

The platform was a mature cyber risk analytics product with a Fortune 500 client base. Three problems surfaced as it matured. Interpreting the data required expert-level knowledge, a pattern common in regulated, data-heavy categories such as banking, finance and compliance. Most users opened the tool once a quarter to produce periodic reports, so it was not part of their daily workflow. And new clients needed roughly 45 minutes of guided setup with active customer support before they could use the product on their own.

When conversational, agentic AI became a market expectation, the client asked us to bring that interaction model into the customer panel. The hard part: the same interface had to serve three very different users. A board member needs a screenshot that explains itself. A CFO needs the narrative behind the numbers. A CISO needs conversational depth to investigate and act.

Our Approach:

The core product question was how to introduce an LLM into a classic dashboard without forcing users to choose between two modes of working. We did not build a separate chat tool next to the platform, because that leaves users with two parallel interfaces. We chose a hybrid: click-and-go for what users already know, ask-and-go for everything else. The hybrid protects the learning curve; users are introduced to AI collaboration gradually, with safety rails at each step.

At the heart of the portal is a four-layer model, with the AI agent (Aria) available at every level:

Layer 0 – Executive: What's happening? 

The first view after login. Key cybersecurity status indicators, classic dashboard, no AI call. Buttons under each metric open prerendered answers built from existing data, instantly. This is the screenshot that explains itself.
 

Layer 1 – Financial: Why is this happening? 

The investigation layer for CFOs and finance leads. Each metric can be explored through prerendered content or through the embedded AI chat. Contextual prompt pills carry complex prompts underneath, so users get expert-level questions without writing them.

Layer 2 – Execution: Do it for me. 

The action layer for CISOs and security operators. Buttons become commands: Create, Identify, Prepare. The agent plans multi-step work and carries it out through its tools instead of describing what the user should do next.
 

Layer 3 – Open Inquiry. 

A free prompt field, always available, for users who know exactly what they want.
The ladder is deliberately not linear. In practice Layer 1 loops back on itself many times (explore → explore deeper → act), because every user has their own decision path.
 

The Solution

Aria is embedded AI native to the platform. It lives inside the customer's interface, brand and authentication boundary, with the analytical depth of the original platform behind it and a conversational front end on top. Users do not switch to a separate tool. The analytics product that used to require expert interpretation became self-serve: users ask, explore and act in one place. Risk is no longer only displayed; it is addressed in the same panel.