December is when most people say they are “thinking about next year” but are really triaging board decks, audits, and family logistics. This briefing is for the small set actually resetting the game board — founders, investors, and operators who want one last upgrade before the year closes.
December rewards people who are willing to zoom out while everyone else is reacting. Most companies default to two modes: churn out year-end reports, then promise a fresh slate “sometime in January” when everyone is back online.
This briefing is a filter. Not predictions, not vibes — just a handful of ideas with real operational consequences for how you hire, allocate capital, and decide what not to do in 2026. If you only have one sitting to think clearly about next year, this is designed to be enough.
Most teams spent 2025 racing to “ship something AI-ish.” This is the argument you keep for the next roadmap review: how statistically fine models become operationally dangerous once they touch workflows, support queues, and customer trust.
Capital is no longer free, but the bench of senior talent leaving over-funded, under-focused companies has rarely been deeper. This piece maps the new equilibrium and what it really means to build with a cost of capital that hurts again.
The heroic firefighter phase is intoxicating and, eventually, fatal. This essay walks through the transition from Slack-driven chaos to boring, compounding systems: weekly cadences, written decision logs, and real ownership instead of endless pings.
Everyone claims to have a high talent bar. Very few can describe the actual process: scorecards, backchannel checks that test judgment, and the discipline to walk away from “almost good enough” senior hires.
Remote work is real; so are power clusters. This contrarian take argues that durable networks in tech, finance, and policy still concentrate in a small number of physical hubs — and that ambitious people should be honest about where they need to be.
Imagine two curves over the last 24 months: one tracking products launched with “AI” in the feature description; the other tracking those that still have meaningful usage six months later.
Before calendars vanish into holidays, run a 60-minute session with your leadership team focused on one thing: identifying and sanding down the “sharp edges” your customers hit most often.
You may not need a new “strategy” for 2026 if you are still cutting customers on the same sharp edges you already know about.
The point of a ledger is not to be omniscient; it is to be falsifiable. These are three shifts worth tracking now, so that next December you are grading your thinking instead of your talking points.
If none of this feels true yet where you sit, that is the signal: advantage flows to people who notice the shift before it is consensus.