Happy Thursday,

AI isn’t slowing down—and neither should we. This newsletter is your shortcut to practical ideas and smarter strategies to grow, lead, and adapt as tech reshapes our industry.

To take it one step further, I’ve launched a new YouTube series. Each month, I’m sharing quick, tactical videos showing how advisors can use tools like ChatGPT to save time, think better, and scale more efficiently—starting with the work you're already doing.

Watch my latest episode here

The AI Edge

AI Application:

Perplexity Computer

Overview

Perplexity Computer is an agentic feature inside Perplexity that lets you describe an application you want built — in plain language — and then actually builds it. Not a mockup. Not a wireframe. A working app with a real interface, real data, and real functionality.

What makes this different from other AI tools is the combination: Perplexity's real-time web search capability paired with the ability to spin up functional tools on demand. You're not just asking a question and getting an answer. You're asking for a piece of software and getting something you can actually use.

It's still early. Credits run out, apps aren't always finished, and you'll hit rough edges. But the point isn't that it's perfect — it's that the gap between "I wish I had a tool that did X" and having something in front of you has shrunk dramatically. For advisors who have always had a mental list of things they wished their tech stack did, that's worth paying attention to.

How to apply Perplexity Computer to the business today:

• Build a quick-reference portfolio dashboard — Describe a simple client-facing or internal view that pulls in real-time data on holdings, sector exposure, and top movers across your book, without logging into your full portfolio management system.

• Create pre-call talking points at the holding level — Ask it to build a view that surfaces recent news, driver themes, and catalyst summaries for individual positions so you walk into client calls with context, not just numbers.

• Prototype a client-facing report — Sketch out what a simplified portfolio summary might look like for a specific client segment, then use that prototype to brief your team or your tech vendor on what you actually want built.

• Test a workflow idea before committing to it — If you've been thinking about a new process or data view but weren't sure it was worth building, describe it to Perplexity Computer and see if a rough version validates the concept before you spend real money on development.

• Build small single-purpose tools for recurring tasks — Think one-screen apps: a quick sector exposure checker, a watchlist with driver tags, a meeting prep summary. Start narrow, keep it simple.

Advanced:

• Connect it to external APIs for live data — Use Perplexity Computer as a starting point and then integrate APIs from your portfolio management system, market data providers, or news feeds to replace the sample data with real client data.

• Create a multi-family office view layered by relationship — Describe a dashboard that groups portfolios by household or relationship tier and surfaces themes driving performance across the whole book, not just individual accounts.

• Use it to pressure-test new service offerings — Before building out a new advisory service or reporting format, build a prototype in Perplexity Computer and test the concept with a trusted client. Real reactions to a real interface are more useful than feedback on a description.

• Chain it with Claude Code for more robust builds — Start with Perplexity Computer to rapidly prototype the concept, then hand the structure off to Claude Code for a more polished, scalable version that isn't burning tokens on every session.

• Build compliance-aware summaries for client communications — Describe a tool that generates draft talking points or position summaries with flagging logic for regulated language, then use your compliance team's feedback to refine the prompt over time.

What could this application mean for the future of our business?

The interesting thing here isn't Perplexity Computer specifically. It's what it represents — which is that the distance between a good idea and a working prototype is now measured in minutes, not months.

For a long time, the advisors and firms with proprietary tools had a meaningful edge. Custom dashboards, internal research views, automated reporting — those things took real development resources to build, which meant they were mostly accessible to large firms. That gap is getting smaller.

What replaces it isn't technology access. It's the clarity of your vision for how you want to work and serve clients. If you know exactly what a better client experience looks like — what data you want at your fingertips, what the pre-call prep flow should feel like, what a great client summary actually says — you can now start building toward that without a development team.

The advisors who will benefit most from tools like this aren't the ones who are most technical. They're the ones who have been frustrated the longest by the gap between what they imagined and what their tech stack delivered. That frustration is now fuel.

See Perplexity Computer in action:

In this demo, I gave Perplexity Computer one prompt and asked it to build a portfolio intelligence dashboard from scratch — real-time driver themes, holding-level catalysts, family-by-family breakdowns. It ran for 19 minutes and burned through my entire credit balance before it finished.

And I still think it's worth watching.

Because what it did build — the performance drivers, the catalyst tags, the per-family portfolio view — shows exactly what this kind of tool can do when you point it at a real advisory workflow. Watch to see what a one-prompt app looks like, where it falls short, and why I'm already thinking about how to rebuild it.

-Matt