What you'll understand by the end of this lesson
- How the ancient memory palace technique works and why it's effective
- How interface design applies location-based memory to navigation and layout
- Why moving a primary CTA disrupts more than just its position
- How breaking location expectations costs comprehension — even when the change looks like an improvement
The principle in plain English
The human brain is exceptionally good at remembering where things are. Our spatial memory is ancient, evolved for navigation, and highly reliable. The method of loci — also called the memory palace — exploits this by associating pieces of information with specific locations in an imagined space.
A memory champion memorising a deck of cards doesn't try to remember abstract values. They place each card as a vivid image in a specific room of a familiar building. To recall the sequence, they mentally walk the building and retrieve each image from its location.
You don't need to be memorising cards for this principle to affect your experience. Every time you reach for a light switch, pour a glass of water, or unlock your phone, you're relying on location memory. You know where the switch is without thinking. That automatic recall is what good interface design builds toward.
A simple example
Your email client has had the same layout for two years. The compose button is top left. The inbox is in the left sidebar. Search is at the top.
One morning, the app updates. The compose button has moved to bottom right. Search is now a floating icon at the top right corner. The inbox is still on the left, but it's now labelled "All Mail."
You know roughly what to do, but you have to think about it now. Every action you performed automatically now requires a small conscious search. The interface hasn't become more complex — it's become less familiar. Your location memory no longer matches reality, and you feel friction that wasn't there before.
Consistent placement builds automatic recall
In web design, consistent placement of key elements across a site — navigation, CTAs, search, logo, contact link — builds the same kind of automatic location memory that the memory palace uses deliberately.
A visitor who has spent three minutes on your site should not have to look for your primary CTA on the next page. It should be where they expect it. If your CTA is at the bottom of every section on your homepage, it should be at the bottom of every section on your pricing page too.
This is not about creativity or visual interest. It's about predictability — which is actually what experienced users want. Predictability means they stop thinking about where things are and start thinking about what they want to do.
Consistency across pages is more important than consistency within a single page. Users scan a page — they tolerate variation on a single view. But between pages, if a key element appears in a different place, it costs comprehension every time. Audit your key CTAs, navigation structure, and trust signals across all pages of your site to ensure they are reliably located.
Why moving the primary CTA matters more than its copy
CRO tests often focus on CTA copy — "Get started" vs. "Start free trial" vs. "Try for free." These tests are legitimate. But an often-overlooked variable is CTA position.
When a visitor has already mentally mapped your page layout on their first visit — even partially — moving the CTA to a different location on a subsequent visit or page forces a reset. Their automatic reach for where the CTA was comes up empty, and they have to consciously search.
This is why radical page redesigns often show a temporary conversion dip even when the new design is objectively better. Location memory built up over time doesn't transfer instantly. Users need time to rebuild their mental map.
If you're running an A/B test that changes CTA position, allow enough time to separate the novelty disruption from the genuine design effect.
When breaking location expectations backfires
There's a temptation in web design to subvert conventions — to put the logo in the centre instead of the top left, to move navigation to the bottom, to bury the pricing link to encourage exploration.
Sometimes this works. More often it costs comprehension, because the visitor's mental map — built from hundreds of other websites — doesn't match your layout. They look for the nav where nav usually is and don't find it. They look for the logo link home and it's not where they expect.
Subverting interface conventions is a high-risk move. Visitors bring a location map from every site they've used before — conventions like top-left logo, top-right navigation, bottom-left footer links, floating search. Breaking any of these creates a moment of disorientation that must be justified by a significant benefit. If you're breaking a convention, ask: what does the user gain from this, and is it worth the location memory cost?
The CRO audit
Look at your site and ask:
1. Are your primary CTAs in the same position across all key pages?
Check your homepage, pricing page, product page, and blog posts. Is the main CTA in the same relative position on each? If it moves between pages, users who are building a mental map of your site will experience friction every time they look for it.
2. Does your navigation appear in the same location on every page?
Changing the navigation structure or position between sections of a site creates cognitive overhead. Users should be able to reach any part of your nav without looking at it — like a light switch they've used a hundred times.
3. Have you run a page redesign recently? Are you accounting for location memory disruption?
If you've recently moved key elements, expect a period of reduced performance as returning visitors rebuild their mental map. This is normal — but if you haven't accounted for it in your CRO analysis, you may incorrectly attribute the dip to the design rather than the transition.
A SaaS company runs an A/B test: Version A keeps the 'Start free trial' CTA in the top-right corner. Version B moves it to the centre of the hero section, where it's more prominent. Version A outperforms Version B in the first two weeks, then the results converge. What most likely explains the initial advantage for Version A?
Location memory helps users find things automatically. But what about guiding users to do things they've never done before? How do you move someone from a simple first action to a complex, habitual behaviour — without losing them along the way?