What's changing right now. Playbooks explain the system. Signals track what's shifting — in private markets, wealth structures, and founder behaviour — so you can move before the crowd notices. Published weekly. Pattern recognition for founders managing their own capital.
Founders copy other people's wealth setups and find they don't fit. A minimum viable family office is five jobs in the right order, and the first is a page of writing that half of them never produce.
Founders count the pension and the index fund as the diversified side of the balance sheet. With the top 10 stocks at 44% of the S&P 500 and the seven biggest, all tech, at about a third, that money is often the same bet as the business.
The story going around is that AI finally lets a founder run their own money and skip the wealth manager. It has it backwards. AI is worst at the actual investment decisions, and best at making the people who make them cheaper to hire.
A full in-house family office runs $700k to $1m a year, which makes little sense below $100m. The cheaper route is to buy the same work from a multi-family office or an outsourced CIO, for a fraction of the cost.
Taking money off the table doesn’t mean you’ve stopped believing in your investment. It’s a strategic move that reflects confidence while also securing some gains.
The founder who sold the company can't sit still. Still checking the portfolio, still eyeing the next deal. The job changed the day the company sold, but the instincts didn't, and that gap is where a lot of newly liquid founders lose money.
Around 30,000 Britons left the UAE between February and April. Most can't afford to come home. The April 2025 UK tax reset closed that fallback, and the structural lesson sits underneath the news. It's older than this year's stress test.
Two family-office reports landed this month with the same quiet finding: the offices adopting AI are using it for paperwork, not portfolios.
Family office direct investing more than doubled in 2025 to $12.9bn, with healthcare second only to AI. On Sunday, Citi named — in its own corporate language — which part of the adviser job AI now does. Two reorganisations in the system around founder wealth, happening at once.
AI is being sold as the operator that wasn't there before. Yale's refreshed search-fund data — out last week — says EBITDA improvement was always the missing piece, and most historical returns came from multiple expansion, not operator skill. The pitch is louder. The math hasn't changed.
BADR jumped to 18%, carried interest to 47%, BPR/APR softened to £2.5m. Three weeks on, the most common excuse for deferring wealth-architecture work is gone. Peer-reviewed research on founder identity arrived alongside, pointing the same way.
SpaceX filed for a $1.75 trillion IPO. Cerebras followed with its S-1. But two other stories in the same ten days — private credit gates at four major funds and a record $225 billion secondaries year — matter more for most founders holding concentrated paper.
