Interview | Sean Morgan: From Data to Decisions — How Lyra Is Changing Private Lending Intelligence

May 15, 2026.

The private lending industry has never had a shortage of data.
The real challenge has always been transforming that data into actionable intelligence fast enough to create an advantage.

For years, meaningful analysis required dashboards, exports, spreadsheets, analysts, and technical workflows that often slowed decision-making. But after recently testing Lyra — Forecasa’s AI-powered analytics assistant — one thing became immediately clear:

This is not simply another AI chatbot layered on top of a database.

Lyra feels more like an intelligent lending assistant trained specifically for private lending workflows — capable of analyzing borrower activity, identifying opportunities, assessing exposure, generating targeted lead lists, and even helping teams forecast future market activity through conversational interaction.

What impressed me most was not only the speed, but the depth of customization.

Within seconds, I was able to generate highly specific borrower searches, identify active lenders in precise markets, evaluate transaction activity, and create actionable lead lists enriched with real lending intelligence.

To better understand how Forecasa developed Lyra — and how lenders are already using it — we spoke with Sean Morgan, CEO & Co-Founder of Forecasa.

Uriel Fleicher: Sean, first of all, thank you for taking the time. I had the opportunity to test Lyra myself, and honestly, what surprised me most was how conversational and intuitive it felt. Why did Forecasa decide to build an AI tool like this in the first place?

Sean Morgan: Great question. Our company has always been very AI-forward. Our Chief Data Officer, Bobby, had been talking internally about AI advancements for the last 18 to 24 months. But the real trigger came from one of our clients.

They basically said:
“Sean, I just want to ask your system questions and get answers back.”

At first, I thought that sounded unrealistic. But after speaking with our development team, we realized this was absolutely something we should build.

So we launched a beta program with a small group of existing clients. We spent months understanding their workflows, the questions they wanted answered, and how they actually interacted with data. That collaborative process eventually became Lyra — our AI-powered analytics assistant.

And since launching it publicly a few months ago, we’ve continued improving it constantly based on client feedback.

Uriel Fleicher: One thing that immediately stood out to me is that Lyra doesn’t feel like a generic AI tool. It feels trained specifically for private lending.

Sean Morgan: Exactly — and that’s the key difference.

A normal large language model may understand language, but it doesn’t necessarily understand private lending workflows.

Lyra has been trained on our entire data ecosystem. It understands concepts like assignments, mortgages, pre-foreclosures, releases, borrower exposure, lending activity, and transaction relationships.

So instead of forcing users to learn complicated workflows or technical filters, they can simply ask questions conversationally.

You don’t need to know coding.
You don’t need to be a data expert.
You simply ask the question you want answered.

Uriel Fleicher: One of the examples I tested was generating highly targeted borrower lists in specific Florida markets based on loan sizes and lending activity. That level of customization was extremely impressive.

Sean Morgan: Yes — and that’s one of the core use cases today.

Some workflows already existed in our platform before Lyra, but now they’re dramatically enhanced. Tasks that previously required multiple filters, exports, or manual analysis can now happen almost instantly.

But more importantly, Lyra also enables workflows that simply weren’t possible before.

For example, one major capability we developed is what we call the “loan lifecycle.”

Uriel Fleicher: What exactly is the loan lifecycle?

Sean Morgan: Historically, it was difficult to connect all the pieces of a transaction over time.

A deed.
A mortgage.
An assignment.
A release.
Potential distress filings.

Those pieces often lived separately.

Now, Lyra can connect them together and understand the entire lifecycle of a loan — from the initial deed through payoff or disposition.

That allows lenders to evaluate things like:

• Outstanding borrower exposure
• Active vs. closed loans
• Ownership history over time
• Current lending relationships
• Concentration risk

That level of contextual understanding becomes extremely powerful for underwriting, business development, and portfolio management.

Uriel Fleicher: And it also goes beyond Forecasa’s internal data, correct?

Sean Morgan: Correct.

Lyra combines Forecasa’s proprietary transaction-level intelligence with external public data sources and third-party integrations.

So yes, we can analyze lending activity, pre-foreclosure filings, market trends, and borrower relationships from our own datasets — but Lyra can also scan public sources for things like:

• Bankruptcy filings
• Litigation
• Regulatory issues
• NMLS-related information
• Broader market intelligence

That creates a much more comprehensive borrower and risk profile.

Sometimes the most important risk indicators aren’t visible inside recorded transaction data alone.

Uriel Fleicher: One thing I noticed is that this can function as both a micro and macro analysis tool at the same time.

Sean Morgan: Exactly.

You can go extremely granular — looking at specific borrowers, loans, lenders, or geographies.

Or you can zoom out and analyze broader lending trends across markets, asset classes, or time periods.

Some users are focused on lead generation.
Others are focused on market intelligence.
Others are focused on exposure analysis or identifying future opportunities.

Lyra can operate across all of those levels.

Uriel Fleicher: Are you seeing particular departments adopting Lyra faster than others?

Sean Morgan: Definitely.

Sales and origination teams are heavily using it for business development and lead generation.

Underwriting teams are leveraging it for exposure analysis, borrower research, and comprehensive risk assessments.

But interestingly, one of the biggest increases we’ve seen is from the C-suite level — CEOs, CROs, CCOs, and executive leadership teams.

Those users often don’t have time to work directly inside complex data platforms. But now they can simply ask questions conversationally and get actionable insights immediately.

For example:

• “What does my competitive landscape look like in this market?”
• “Which lenders are pulling back?”
• “Where should we expand next?”
• “What markets show growing opportunity?”

That accessibility has significantly increased executive engagement with the platform.

Uriel Fleicher: Another thing that surprised me was that Lyra can also generate charts, graphics, and even help structure campaigns.

Sean Morgan: Yes — because we wanted it to become more than a search tool.

You can ask Lyra to create graphics, summarize trends, build targeted borrower lists, and even assist with outreach workflows or campaign ideas.

And once users create workflows they like, those workflows are saved permanently.

So over time, every user essentially builds their own personalized AI assistant trained around their specific role and objectives.

Uriel Fleicher: That’s actually a very important distinction. It doesn’t feel like users are simply accessing a tool — it feels like they’re building their own intelligence layer.

Sean Morgan: That’s exactly right.

Over time, Lyra understands more about the user, their role, and the types of workflows they care about.

An underwriter may see completely different workflows than a business development executive or a capital markets professional.

We also built prompt libraries and workflow templates because we realized many users were new to AI and didn’t necessarily know what questions to ask.

So we guide users based on their role and continue improving those workflows every week based on how clients are using the system.

Uriel Fleicher: One thing you mentioned that I found fascinating was the idea of “agent-to-agent” communication. Can you explain that?

Sean Morgan: Absolutely.

We believe the future is not just people interacting with AI agents — it’s AI agents interacting with each other.

For example, in the future, your CRM agent could communicate directly with Lyra. Your underwriting systems could communicate with external data agents automatically.

That’s why we built compatibility through our MCP server integrations with tools like Claude and other AI ecosystems.

The future is about interconnected intelligence systems that can collaborate and act proactively.

Uriel Fleicher: So eventually the system becomes predictive — not just reactive.

Sean Morgan: Exactly.

Today, most systems analyze what already happened.

The future is about forecasting what is likely to happen next.

How do we identify opportunities before competitors do?
How do we predict borrower activity?
How do we anticipate market shifts earlier?

That’s where we believe AI-powered intelligence is heading.

Uriel Fleicher: One last question. Does Forecasa track loan maturity dates as well?

Sean Morgan: Yes — for a large percentage of loans.

That’s actually one of the most valuable workflows today because lenders want to identify upcoming maturities before borrowers actively begin shopping for capital.

Timing matters tremendously in private lending.

If you can identify a borrower before they need financing, that creates a major competitive advantage.

Uriel Fleicher: Sean, thank you again for taking the time to share these insights. After testing Lyra personally, I genuinely believe this represents a major shift in how lending teams interact with data.

This isn’t simply about accessing information faster.

It’s about transforming raw data into real-time intelligence, actionable opportunity, and ultimately better decision-making.

And in a market where speed, timing, and insight increasingly define competitive advantage — that shift matters.