I’ve lost count of how many times I’ve done this: standing like an idiot before a door with a handle screaming “PULL ME!” while a tiny sign says “push.” We all have. Don Norman, the User Experience (UX) legend, labeled this a “Norman Door”: design so bad it needs instructions to use.
And I’m here to tell you: Your product might be a Norman Door.
In my years as a product manager, from enterprise to small business software, I’ve cringed watching users struggle with prototypes my team swore were intuitive. This door problem isn’t abstract design theory. It’s a product disease that’s bleeding your conversion rates dry.
Related: The 10 Obstacles Keeping You From Great Product Design
Affordances aren’t just design jargon
When I first learned about affordances, visual cues showing how something works, it seemed like design-school fluff. Then I watched a user test where people couldn’t find our “obviously placed” save button, and suddenly Don Norman didn’t seem so theoretical.
Take Robinhood’s original interface. While dinosaur brokerages buried users in cavernous websites, Robinhood stripped investing to its essentials: up means good, green equals money, swipe to act. It wasn’t just “easier.” It fundamentally changed who wanted to invest by making actions obvious.
The mistake product managers and teams make is designing for ourselves. Most products make perfect sense if you built it and already know how it works. But customers only have what is in front of them, and that needs to be intuitive.
Your customers are always right
When users struggle with your product, it’s tempting to blame them. “They didn’t read the manual.” “They need training.” “Not our ideal customer persona.”
This is the product version of slapping a “PUSH” sign on a door with a pull handle. Good design doesn’t need a manual.
Bad error messages are my personal pet peeve here. Many are written by developers for developers: “Error code 5432: Null pointer exception in transaction handler.” Great, helpful.
HCI research (and common sense) shows that effective error messages need to explain what happened in human language, clarify consequences and suggest a fix. When Slack tells you “You’re trying to upload a file larger than 50MB” and immediately suggests “Try compressing it or using our Google Drive integration,” they’re preventing the rage-quit I’ve seen too many times in usability tests.
Related: How Prioritizing UX Design Can Fuel Long-Term Growth in the Next Decade
Constraints as a business strategy
Industrial designers use constraints intentionally — like how a SIM card only fits one way (though I still somehow get it wrong every time). Product constraints aren’t limitations; they’re clarity engines. The most successful products I’ve worked with deliberately limit options to prevent catastrophic errors and the blank stare of cognitive overload.
Look at how Figma entered the design tool space. Rather than trying to cram in every Adobe feature accumulated since 1990, they constrained their tool to the essential components of interface design. They deliberately avoided complex layer effects in favor of making collaboration seamless. I watched countless designers switch over once they realized these “limitations” actually sped up their workflow dramatically.
For your business (and mine), this means ruthlessly killing features that don’t support your core value. Features that don’t align can actively undermine your UX.
The feedback loop that actually matters
Don Norman’s “gulf of evaluation” — the gap between what users expect and what happens — applies as much to your quarterly business reviews as it does to your checkout flow.
When a user takes an action in your product, how quickly do they know if it worked? If the answer is “they have to check email” or “they’ll find out later,” you’ve created an evaluation gulf that will drown user confidence.
I’ve been guilty of this. My team once built an “instant” data export feature that actually took 30 seconds to run — with zero feedback during processing. User testing revealed people were clicking the button 5-6 times, thinking it was broken. We added a progress bar, and users now wait patiently.
Amazon mastered this with one-click ordering. The moment you tap “Buy Now,” you get confirmation that your order is processing. It resolves the uncertainty anxiety that kills conversion rates.
For your product strategy, this means investing as much in confirmation states as feature development. Users remember uncertainty more than they remember your feature list.
Related: What a Complete User Experience Process Looks Like — and How Investing in One Can Benefit Your Business
Apply these principles now
These HCI principles are practical business tools I use daily. Here’s how to apply them immediately:
-
Find your product’s “Norman Doors” by watching new users interact without help (and resist the urge to explain when they struggle)
-
Map the entire user journey, flagging every moment of confusion or surprise
-
Review error states with the same rigor you apply to feature development (seriously, audit them all)
-
Evaluate feature requests against your core constraints instead of automatically adding them to your backlog
The products that win aren’t those with the most features or flashiest tech. They’re the ones that communicate clearly, prevent errors gracefully and deliver feedback instantly. They’re doors that you know how to open the first time.
Though honestly, those “PUSH” and “PULL” sign manufacturers are probably doing just fine. There’s always money in bad design.
