Notes from building software and a business.
I write about the decisions behind products: what to build, what to leave alone, how teams work, and what happens when an idea meets the realities of a business.
These are perspectives shaped by my own projects and experience, alongside questions I'm still working through.
Latest articles
- Sep 21, 2026 product
AI Made Me Faster. Now I Have to Be More Careful About Direction.
I had told the AI where the screen belonged, how the queue should pace delivery, what the preview should show. I had forgotten to mention the timezone. By the time I reviewed it, the omission was already a default in working code.
- Sep 17, 2026 product
The Business Rules Are the Product
A salesperson could prepare a proposal, negotiate it, and take it through to a sale. At that point, I took away their permission to edit it. Where that restriction sits is the product.
- Sep 16, 2026 product
A Working Product Is a Very Convincing Way to Avoid Selling
I postponed outreach for Zaplink because I thought larger customers would want their own WhatsApp number. None had asked. A customer you haven't spoken to can have a surprisingly detailed requirements list.
- Sep 15, 2026 product
Before You Automate a Process, Watch Someone Do It
The first proposal builder I made for The Woods looked like the proposal itself. I eventually removed it. "Prepare a proposal" is a reasonable description of a task, and nowhere near enough to build the thing that does it.
- Sep 14, 2026 product
The First Version of Your Product Includes You
One click creates a client's deal stages, properties, and workflows. Then we check the setup fits the team that will use it. The customer's experience includes the work you do around the product.
- Sep 10, 2026 product
Your Biggest Competitor Might Be a Spreadsheet and a WhatsApp Group
At The Woods, every customer file had its own WhatsApp group and a proposal could take a day to assemble. Before you replace someone's spreadsheet, ask what moving would cost them.
- Sep 9, 2026 product
When You Can Build Almost Anything, What Should You Build?
I nearly built a marketplace for personal trainers because I didn't want to pay for a workout app. What stopped me was never whether I could build it — it was the work that would have arrived afterwards.
- Apr 14, 2026 insurtech
The Anatomy of an MGA Tech Stack at $50M, $100M, and $250M in Premium
The technology decisions an MGA makes in its first two years will either carry it to $100M or force a rebuild at $40M. This piece maps the actual systems, costs, and architecture patterns at three stages of growth, with 19 sources cited.
- Feb 4, 2026 engineering-leadership
The Delivery Illusion
Why busy teams feel stuck, and how to measure what actually moves the product. The gap between output and outcome is expensive precisely because it looks like everything is working.
- Jan 28, 2026 engineering-leadership
Why Your Team Slowed Down
Teams don't slow down because people get lazy. They slow down because well-intentioned checks compound, until preventing mistakes costs more than the mistakes did.
Shipping Notes
A weekly roundup of what I've written and built, plus useful things I'm reading or exploring. The full articles live here; the email helps you catch up.