Notes on building the AI-native web.
Page 17 of 17
The AI pilot that never ships
It demos well, everyone is pleased, and eleven months later it is still a pilot. The reasons are organisational, not technical.
What to automate in testing, and what to check by hand
Chasing total coverage produces a slow suite nobody trusts. A small, fast, meaningful set plus deliberate manual checks beats it.
Designing the interface for when the AI fails
Most AI interfaces are designed for the successful case. The failure states are where users decide whether to trust it again.
Environments: how many do you actually need?
A staging environment nobody trusts is worse than none. What each one is for, and the property that makes it useful.
How often should you update a page for AI search?
Changing the date is not updating the page. What freshness actually signals, and which pages genuinely need it.
Do press releases still do anything?
As a link-building tactic, no. As a way to get accurately described at credible sources, occasionally yes.
The 30-minute AIEO audit
Six checks you can run on your own site this afternoon, in the order that finds the most damage fastest.
The questions to ask before any AI project
Eight questions that take an hour to answer and predict, more reliably than anything else, whether the project will ship.