I launched NegotiatorIQ on Tuesday.
No PR firm. No countdown timer. Just me refreshing analytics like a crazy person waiting for the site to crash.
It didn’t crash. Which honestly felt like the first win.
Here’s how we got here.
It started with a simple frustration. I’ve taken DISC. Myers-Briggs. StrengthsFinder. They’re great for understanding personality. But when I sat across the table in an M&A deal, a federal contract negotiation, or an enterprise sales call — none of that translated. Knowing I’m an “INTJ” didn’t help me close. Not once did I think “well, as a high-D, I should anchor first.”
So we built our own assessment. Four negotiation styles based on your natural tendencies — not your personality type. And unlike those other assessments, we believe you can actually improve on your blind spots. That’s the point.
Beta testing went well. Users loved it. I started thinking we might actually know what we’re doing.
But something was missing. The biggest problem in negotiation isn’t talent. It’s that there’s no common language. No repeatable process. Just jargon and instinct. So we developed the 8 Moves of Negotiation — a framework that takes you from Know Yourself through Run the Numbers, Set Your Strategy, Control the Opening, Control the Clock, Signal, Redirect, and Close the Deal. Moves 1–4 are preparation. Moves 5–8 are execution. Simple. Repeatable. Trainable.
Then I needed a launch concept. Something that would let me use both my backgrounds — business and cybersecurity. That became the NegIQ-234: an analysis of 234 real ransomware negotiation transcripts. Over 11,000 messages. 24 criminal gangs. Turns out, the tactics criminals use at the table are the same ones that show up in every boardroom — anchoring, time pressure, emotional manipulation, reframing. It was the perfect blend of both worlds, and it gave us real data instead of theory.
Great. Ready to launch.
Not quite.
Because apparently we hate sleep, we decided to add AI. We built a Crisis Mode and a Ransomware Simulator — both powered by a model we actually fine-tuned ourselves. Not just prompts. We used Llama 3.2 (3B parameters), LoRA fine-tuning, and AWS SageMaker for ML ops. It’s been a journey. Our simulator still occasionally develops ethics and refuses to do the simulation. We’re working on it.
Meanwhile, we submitted about 30 applications to VC funds and incubators, including Jumpstart here in Cleveland. First round was May 2025. The feedback: “Good idea, but you need more than an idea.”
So I did what every founder does after 30+ rejections. I spent the summer by the pool. With my laptop. Pretending to relax while cranking out an entire library of training videos.
We submitted again in November. Jumpstart. Made it to the final round — 120 applicants down to 30. And we missed that one too.
This time I didn’t sit by the pool. I rolled up my sleeves and started coding. Hard. Probably broke as many things as I fixed.
Somewhere in all of this, I picked February 17th as the launch date. No particular reason — I was just drawn to it. Turns out February 17th is Lunar New Year and the start of the Year of the Fire Horse — a rare 60-year cycle representing high energy, rapid progress, and independence. Bold action, passion, and yes, volatility. If that’s not a startup launch, I don’t know what is.
Reid Hoffman once said if you’re not embarrassed by your first product release, you launched too late. Launch as soon as you have a “quantum of utility” — the minimum functionality that lets at least one person solve a problem they couldn’t before.
We’re there. It’s not perfect. There are things I already want to rebuild. But it’s real, it’s live, and actual humans are using it — and so far nobody’s asked for a refund on a free product, so I’m counting that as traction.
The journey isn’t over. It’s just starting.
Thanks for being part of it.
