Explore for more
↓
The Cognitive Budget Model
Attention is already divided
Users don’t arrive with their full attention. Notifications, unfinished thoughts and competing decisions have already taken a share.
A clear interface helps them see what matters without adding more work.
01 — Entry State
Users don’t start from zero.
They start from noise.
Notifications. Context switching. Half-finished decisions.
Design begins by respecting that.
02 — Signal
Primary intent must be obvious.
If users need to interpret the interface, you’re already spending budget.
Good design reduces interpretation.
It doesn’t add explanation.
03 — Load
Every badge, secondary CTA and decorative element competes for attention.
Every extra choice adds to the cognitive load.
Complexity is often additive.
Clarity is subtractive.
04 — Outcome
When cognitive load drops:
• Decisions become easier
• Actions become more confident
• Fewer things need explaining
The product becomes easier to use without becoming less capable.
I use this model to keep products understandable as they grow.
It helps decide what needs attention, what needs explaining and what can stay out of the way.
That’s a big part of what I mean by Product Storytelling.
Why is making things simple so effin hard?
Because most bad decisions start with a perfectly good reason.
Features, options and exceptions rarely get there by accident. Someone needed them, promised them or thought they might be useful later. Removing them means making trade-offs, and that's where things get dicey.
Cognitive load grows when
“This customer needs it.”
“Sales promised it.”
“It might be useful later.”
“Removing it could upset someone.”
Cognitive load drops when
“We don’t need this anymore.”
“That can wait.”
“No, we’re not adding that.”
“This is the way most people use it.”
Field Notes
Honestly, these notes from real projects don't always make me look good. That's partly why I wrote them. You deserve to know who you're considering working with.
One person can do an unreasonable amount of work now
Two years ago a friend of mine was burned out and certain he’d never go back to work.
When I asked recently how he was doing, he started listing things he’d made, faster than I could follow. Most of them were already running.
AI is very good at doing the wrong thing
AI can follow a bad brief with impressive speed.
I’ve found it most useful when I treat the first answer as something to question, not something to ship.
A simple screen can take a lot of arguing
Old web portals tried to fit the whole internet onto their homepages. Google offered a search box and a way out.
Cognitive load isn't one UX problem
It shows up in different places depending on the product.
These are some of the places I tend to look first.
Onboarding
New users don't know the product yet. I look at what we're asking them to understand before they've had a chance to see the value.
Dashboards
Teams know what every number means. Users usually don't. The question is what deserves attention first — and what can wait.
AI Features
AI can do a lot, which makes it tempting to expose a lot. I look for the shortest path between what the user wants and what the AI can actually help with.
Settings Architecture
Flexibility adds choices. Sometimes that's necessary. The work is figuring out which decisions belong to the user and which the product should make for them.
Design Systems
Consistency helps, but a perfectly consistent interface can still be confusing. Components need to serve the hierarchy, not determine it.
Product Growth
Products accumulate features, edge cases and old decisions. I look for places where yesterday's good reasons are making today's product harder to use.