How I Used AI to Brainstorm, Validate, and Build the Foundation of Savvyn Insights
- Catherine Del Vecchio Fitz
- Mar 25, 2025
- 5 min read
Ideas Are Easy. Execution Is the Challenge.
Once I committed to building Savvyn Insights, I had to answer a critical question:
What exactly am I building?
I knew three things:
1️⃣ I didn’t want to build a business that scaled by adding more hours. It had to be flexible, high-leverage, and built for outcomes—not just effort.
2️⃣ I wanted to use my cross-functional background in science, medicine, data, and operations to create real impact. Not surface-level wins. Not performative progress. Actual, meaningful change—with integrity at the core.
3️⃣ It had to work for me—not the other way around. I’ve done the grind, and I wasn’t about to build something that burned me out in a different outfit.
So began the brainstorming phase: a chaotic blend of genius-level strategy and just the right amount of unhinged.
Step 1: The Brainstorm Phase
Before refining the model, I dumped every possible idea onto the table—from the practical to the absolutely ridiculous.
Some ideas were solid. Some were...interesting. Some made me question my life choices.
Here’s a glimpse into what made the list:
💡 AI-powered content generation for scientific publications → Sounds useful. But did I really want to become the Oncology Abstract Machine? No thanks.
💡 An AI-driven system for structuring messy datasets into usable insights → This one stuck. Because weirdly, I love this work.
💡 Helping startups make sense of their own data before it buries them alive → Low effort, high value. A very real sweet spot.
💡 AI-powered biotech intelligence platform → A SaaS product that structures biotech and oncology market intelligence in real-time. Promising, but a heavy lift.
💡 An AI-driven platform to drive adherence to MRD-based post-treatment cancer surveillance → Clinically meaningful. Technically feasible. But required provider-facing infrastructure and regulatory lift I wasn’t ready to take on (yet).
💡 AI-assisted curation of real-world datasets for oncology applications → Technically niche, but high-value. The challenge? Most people don’t realize how messy real-world oncology data is until they’re already drowning in it.
💡 A clinical decision support layer that bridges NCCN guidelines with emerging diagnostics → Loved the concept: practical, scalable, and genuinely helpful. But trying to keep up with every new diagnostic while avoiding a regulatory rabbit hole? Not the dream.
💡 Founder’s AI-powered playbook → Scalable, yes. But do I want to be the "Startup Whisperer"? TBD.
💡 AI-enhanced Chief of Staff consulting → Would people just think I’m a sentient productivity tool? Pass.
💡 A subscription-based AI assistant for precision medicine strategy → Too close to advisory disguised as product. Tempting, but not it.
At one point, I had the thought: "What if I just become a human ChatGPT for research teams?" …which was quickly followed by: "That would be a special kind of exhaustion."
So, some ideas got cut pretty fast.
Step 2: The ‘Go Big or Go Home’ Phase (a.k.a. ChatGPT Went Off the Rails)
I let AI help me brainstorm. Big mistake—or big inspiration?
Instead of suggesting normal business models, ChatGPT decided I should either:
1️⃣ Assemble an AI-powered oncology SWAT team OR
2️⃣ Build a shadow empire of AI-driven businesses and rule precision medicine like a data-loving overlord.
…Which, honestly, was a good reminder that I still need my brain—and humans generally—in the mix.
Here’s what ChatGPT tried to make happen before I pulled it back to reality:
1️⃣ The “Savage AI-Driven Oncology Skunkworks”
🚀 What if I built an invite-only, AI-powered SWAT team of experts solving high-impact oncology problems FAST? No bureaucracy. No Zooms. Just a strike force of elite minds solving impossible problems. Biotech and oncology orgs apply to get my team’s help—but I only take on the most intellectually exciting projects.
💡 Why ChatGPT Thought This Was Genius:
✅ Feels like a secret mission instead of a job.
✅ High-impact + high-intensity = ultra-satisfying.
✅ Solving cancer problems, but make it elite.
🙅🏻♀️ Why I Had to Shut It Down:
❌ I like my sanity.
❌ This sounds like an Avengers recruitment process.
❌ How do I even explain this to clients? (“Apply for my secret society?”)
2️⃣ The “AI-Powered Shadow Empire” for Oncology & Precision Medicine
👑 What if I didn’t build ONE business—but an entire network of ultra-efficient AI-powered oncology ventures? Instead of one company, I’d launch a portfolio of AI-driven businesses that generate revenue while I sleep.
💡 Why ChatGPT Got a Little Too Excited:
✅ Diversified revenue—no single point of failure.
✅ Automate everything, own multiple revenue streams.
✅ Could be built asynchronously—no meetings, no stress.
🙅🏻♀️ Why I Had to Shut It Down:
❌ AI thinks I’m an immortal productivity machine.
❌ “Just launch a biotech empire” is not practical business advice.
❌ I would spend the next decade managing infrastructure instead of actually doing what I love.
Step 3: Refine & Validate With AI (For Real This Time)
Once I cleared the clutter and reined in ChatGPT, it was time to validate the remaining ideas. I needed data to make informed decisions, so I used AI strategically to refine a real, scalable business:
Market demand – AI helped scan trends in oncology, AI-powered consulting, and research-driven businesses to see what’s growing vs. what’s just hype.
Competitor analysis – Looked at similar offerings to see if they were actually good or just noisy.
Pricing structures – Tested pricing structures, from one-time services to subscriptions and retainers.
Real pain points – Turns out messy data is still ruining productivity across biotech and research orgs.
That clarity was the turning point.
Step 4: The Final Model Takes Shape
As the dust settled, this was clear:
Savvyn Insights helps early-stage teams and high-performing individuals structure messy data, optimize workflows, and harness AI—without the bloat of enterprise systems.
It’s lean, scalable, and built around what I do best—translating chaos into clarity.
By focusing on AI-driven data curation and workflow optimization, I could scale the business while maintaining flexibility.
What this looks like in practice:
✔ Clients pay for structured data and automation—not my hours
✔ AI handles the repetitive grind; I handle the strategy and oversight
✔ I stay focused on oncology, biotech, and data-heavy teams where the value is real
✔ I work flexibly, sustainably, and efficiently—without burning out
Which is much better than what I originally wrote: “I take your disaster of a dataset and make it make sense. You’re welcome.”
Still true. Just... more polished now.
Step 5: Vision & Purpose Lock In
With the business model now taking shape, the vision and purpose of Savvyn Insights came into clear focus:
Vision
To build a high-impact, AI-first consulting business that empowers organizations and individuals—starting with life sciences and expanding into other data-driven industries—to work smarter, faster, and more effectively. We are committed to creating a workplace that aligns with true balance—prioritizing flexibility, autonomy, and sustainability without sacrificing ambition.
Purpose
Everyone should have access to tools that make them more efficient and effective—without needing to be a tech expert. Savvyn Insights exists to bring automation and AI within reach of the people doing the real work.
But more than that—we’re here to prove a few things:
✔ That high performance doesn’t require burnout.
✔ That lean teams can punch above their weight.
✔ That balance and ambition don’t have to be opposites.
Final Thought
Savvyn Insights is my answer to the question: How do I build something smart, scalable, and sustainable—without sacrificing my life to do it, and without compromising the impact I want to make?
It’s not about working harder. It’s not about doing more. It’s about building smarter—with less noise, more intention, and systems that scale.
No fluff, no grind. Just a plan to disrupt the status quo, run things on my terms, and prove that success doesn’t mean sacrificing your sanity.
NO BIG DEAL.
🔜 Up Next: How I Landed on the Name Savvyn Insights
Naming a business is an adventure in itself. My next blog will dive into how Savvyn Insights got its name. Stay tuned!
Comments