Writing for the people who run design and research teams: how the work gets measured, resourced and understood. Plus our research notes on synthetic users.
Start here
Design is not harder to measure than engineering. It is measured against the wrong unit. Why artifact counts fail, why attribution breaks, and what belongs in a business review instead.
Perspectives5 min read
Perspectives5 min read
Headcount is a budget line. Coverage is the decision that determines what your design org can actually do. Why allocation drifts invisibly, and how to see it before the quarter is over.
Perspectives5 min read
Design performance reviews reward proximity and visibility rather than contribution, and everyone in the room knows it. Why the evidence goes missing, and what a better record looks like.
Research notes5 min read
Figma surveyed 8,403 practitioners and found AI made teams measurably more productive without making their products better. The gap is not a tooling problem, it is an operating model problem.
Perspectives5 min read
DesignOps is the function that makes a design organization run, and the average company does roughly a fifth of it. What the work actually consists of, why it gets cut first, and what happens to it when nobody owns it.
How-to6 min read
Ten questions that separate a grounded simulation platform from a prompt with a dashboard, why each one matters, and what a good answer sounds like. Includes the questions we would want asked of us.
How-to5 min read
Most design reviews with leadership are a portfolio walkthrough that answers questions nobody asked. A structure for the quarterly account of a design organization, in the four sections a CFO already knows how to read.
Perspectives6 min read
The three ways a product team gets an answer about a design, what each one actually measures, what each costs in calendar time, and a rule for picking between them on the decision in front of you.
Perspectives5 min read
A plain definition, how a synthetic study actually runs, what the peer-reviewed evidence supports, and the questions it still cannot answer. Written for teams evaluating the category for the first time.
Perspectives5 min read
Teams do not skip research because they think it is worthless. They skip it because it arrives after the decision. What actually blocks research, and what changes when you fix the timing.
How-to5 min read
The honest answer is that no ratio survives contact with your org. What holds up is a way of sizing research against decision volume and risk, and a headcount case built on coverage gaps rather than principle.
Perspectives5 min read
Managers lose weeks a year to performance reviews, and design managers have it worse because the evidence they need was never written down. Where the time actually goes, and how to get most of it back.
How-to5 min read
The claim that design drove a result gets discounted the moment you make it, and usually should be. A practical method for tracing work to outcomes that holds up when someone pushes back.
Research notes4 min read
In Nature, GPT-4 simulations predicted the results of 70 preregistered experiments at r = 0.85, and r = 0.90 on unpublished studies. What this validates, what it does not, and what to do with it.
How-to6 min read
Simulation quality is mostly a grounding problem. A practical guide to feeding synthetic panels the transcripts, tickets, and behavioral data that make their answers worth acting on.