Writing on measurement, evidence and the gap
Between what users say and what they do.
When the user cannot leave: measuring internal tools
Every UX signal you own is a proxy for exit, and an internal tool removes the exit. What an HR system costs instead, and why the perception reading needs a second discount.
The expectation gap: when usable products still lose people
Nothing is broken, the usability scores are good, and they leave anyway. Why usability testing is structurally blind to this gap, and where the promise that caused it was actually made.
Accessibility is a gate, not a score
Averaging a legal obligation into a composite converts it into a preference. Why conformance sits outside the score entirely, and where the experience of using a product under constraint actually lives.
The four currencies of friction
Time to outcome, attempts, assistance, abandonment. How to count each one honestly, the trap that corrupts each, and why they are never averaged into a single number.
Scoring our own product in public
We ran RUCF on this site before launch and published the result, including the parts that came back guessed. A framework that will not score its own author is a marketing asset.
What a scorecard cannot tell you
A self-assessment reports what you tell it, structured and scored. Here is where its usefulness ends, stated by the people who built one.
How to price friction in your own currency
Abandonment times traffic times value per completion. Support contacts times cost per contact. The arithmetic that turns a UX finding into a line item a CFO will read.
The say-do gap, from Samuelson to your checkout funnel
Economics documented it in 1938. Social psychology in 1934. UX measurement still averages it away. A short history of the oldest finding your dashboard ignores.
Loud friction is a receipt, not a discovery
Why most research budgets go to confirming what the support inbox already said, and the sorting question that stops it.
Phantom friction: when rebuilding the interaction is the wrong answer
The flow works, users call it broken, and the standard response spends the budget the real fix needed. The most commonly misdiagnosed state in the discipline.
Silent friction: the failures no survey will find
Users have absorbed the cost or blame themselves for it. The mechanisms behind the most valuable state in the framework, and why it is the cheapest to fix.
Stop reporting average UX. Performance engineering stopped fifteen years ago.
An average experience is a fiction nobody has. What P50 and P10 reveal that a mean conceals, and what a wide tail gap accuses you of.
Your research expired the day you shipped
Evidence describes a product that no longer exists. Deploy decay: why confidence should fall on release, not on the calendar, and how that closes the loop by itself.
Why expert review is the weakest evidence in UX, and why we weight it 0.30
Everyone in the field knows it. No other framework says it numerically. On heuristic reviews, AI classification, and the discount both deserve.
Your users are not telling you what your product costs them
They are not lying; they have stopped noticing. The founding argument of the Real User Cost Framework, in one article.
Reading is slower than measuring
The scorecard applies all of this to your product in about two minutes.