Loading
Skills you'll build
Making complex tools feel simple to use
Helping users understand what AI can (and can’t) do
Fixing broken expectations before they cause confusion
How do you teach people to use something that goes through step-change improvements every 6 months?
Put it like this; there's a clue that ChatGPT isn't as intuitive as a traditional search engine.
It's that companies are hiring for a skill known as "prompt engineering".
In short: the change in complexity of AI is outpacing behavioural evolution—by a lot.
It's harder to build predictable and transparent experiences when even literal computer scientists can't predict the outcome of a prompt, or explain it to someone on the train.
This is a case study about ChatGPT.
But more so, it's about core principles of design that are being stretched and tested by AI.
And I think it explains the strategy for Apple Intelligence.
Written by Peter Ramsey
Published
Last edited
Other popular case studies
Free
8 min to learn...
Why Adding Friction Improved Built for Mars' Conversion Rate by 25%
with aCase Study
focused on
Built for Mars
A behind the scenes look at how I redesigned the Built for Mars sign-up flow, and iterated from early feedback.
Skills you'll build
The two "obvious" fixes that did nothing
Why you shouldn't always force an input
Why limiting the user's choices can be beneficial
Hinge built its reputation on a single promise: it's the dating app that wants you to leave. This is a breakdown of how the interface is actually engineered to make sure you don't.
Skills you'll build
How dating apps keep you engaged (not romantically)
The design equivalent of clickbait headlines
How Hinge leverages social scarcity
BFM+Exclusive
6 min to learn...
Google is Hollowing out Waze, and that's a Problem for Apple
with aCase Study
focused on
The challenge for a navigation app isn't to efficiently show you the best route, it's to make you believe that you've found the best route.
Skills you'll build
How strategy influences design
Waze is cornered, and slowly shrinking
What Apple's strategy appears to be
BFM+Exclusive
8 min to learn...
How To Design A Loyalty Scheme That Actually Motivates People
with aCase Study
focused on
Costa
From rat mazes to coffee apps: why the motivation to reach a reward increases the closer your users get to their goal.
Skills you'll build
How to use the Goal Gradient Effect
Spinning the flywheel of purchases
Why your rules need to be clear
BFM+Exclusive
7 min to learn...
If a Supermarket Scanner Needs Explaining, Something’s Broken
with aCase Study
focused on
Sometimes, the best onboarding is no onboarding, and simple products get bogged down with unnecessary explanations. This case study on Waitrose, Sainsbury's and Asda demonstrates when to just let the user try it for themselves.
Skills you'll build
Sometimes the best onboarding, is no onboarding
“Show, don’t tell” often beats education
Data can’t always tell you when onboarding is wrong
BFM+Exclusive
8 min to learn...
Monzo Nailed the Start of The Savings Challenge (But Not the Middle)
with aCase Study
focused on
Monzo
An analysis of Monzo’s 1p challenge, focusing on the middle slump, where progress slows and motivation quietly fades.
Skills you'll build
Delight can replace progress, but only temporarily
"The middle" is the real retention problem
Motivation needs contextualised reference points