There’s a version of your user that only lives in research decks and design briefs. They’re focused, patient, and unhurried. They’ve read the onboarding copy, they’re sitting at a clean desk, and they have nowhere else to be.
We want them to exist in this way, but the reality is actually the opposite.
This is the argument James Eccleston, a UX designer and founder of Bridge Studio, made in a social media post that sparked a wave of conversation across the UX and product community. And, if you’re serious about user research and user testing, it’s one worth sitting with.
The reality is that your real user is checking your app on a crowded platform with a train pulling in. They have 10 computer tabs open while trying to complete a form. They’re sometimes stressed, mentally elsewhere, and more often than not, not that invested in your interface. They just need it to work.
If you’ve ever launched something that tested beautifully and performed poorly, this article is for you. What follows are three user research principles that will help you close the gap between the user you assume and the human who actually shows up.
The User You’re Designing for Doesn’t Exist
UX research has a focus problem, said James in this episode of our UX Spotlight podcast. Not a lack of it, but too much of it.
Testing conditions are often too clean, too controlled, and too forgiving. You recruit the right users, brief them carefully, and observe them in environments where the only thing they have to do is use your product. Then you’re surprised when the same product fails in the wild.
The gap between the user you test for and the user who actually shows up is where products can be less effective.
Bridging that gap isn’t just a design challenge, but a user research challenge, and it starts with three principles.
3 User Research Principles Worth Building Into Your Practice
1. Cut it in half. Then cut it in half again.
The first instinct when something isn’t working is to add: more guidance, more tooltips, more explanation. However, the better question should be: what can we take away?
Every element on a screen is a small tax on your user’s attention. Every checkbox, every pre-emptive warning, every step that could be automated is a cognitive load you’re asking a distracted human to carry. And distracted humans have very little attention to spare.
Here’s a practical example of how Apple understood this early. Apple Pay reduced one of the most cognitively loaded everyday tasks — paying for something — to a double-click and a glance. No selection screen, no confirmation step, no thinking. Just a reaction. That’s the real benchmark for successful UX: the moment your user stops noticing the interface entirely.
Wise does something similar with financial complexity. It surfaces exchange rates, transfer fees, and delivery times without ever making the user feel like they’re doing mental arithmetic. The information is there and it doesn’t require effort.
James suggests this useful exercise: Take what’s on the screen and cut it by half. Then ask yourself whether it can be cut in half again. If you can, you probably should have the first time. Push until it breaks. And if it doesn’t break, you’ve found what actually needs to be there.
Another example is how a fintech team applied this thinking when redesigning a prepaid card application. They arrived to find a full page of instructions users were expected to read before topping up, such as currency rules, wallet requirements, conditions, all of it front-loaded, and all of it ignored. The solution wasn’t to rewrite the instructions. It was to eliminate the need for them entirely, surfacing guidance only when relevant and removing the options that led to errors in the first place.
The lesson here is simple: Less content leads to less friction.
This is also where unmoderated user testing earns its place, helping to quickly surface unnecessary friction and redundant steps without the overhead of a guided session. It’s the quickest way to see what’s getting in the way.
2. Test for the ‘real world, not the ideal one.
Oftentimes, ‘lab’ testing environments produce ‘lab’ results. The problem is that these results don’t actually factor in for the ‘messy’ reality of everyday life.
Instead, good user research accounts for the conditions your users are actually in and not the conditions you wish they were in. That means designing tests that introduce the kind of friction, distraction, and time pressure that real use involves.
A good approach is to give users a time limit that creates genuine urgency. Ask them to complete a task while walking around. In a moderated session, for instance, let participants get halfway through something, interrupt them, and see if they can find their place again. Test whether your product is usable with someone talking to them at the same time.
These aren’t stress tests designed to break things, but rather think of them as calibration tools; ways of checking whether what performs well in a controlled environment also holds up when your user is nervous, rushed, or doing three things at once.
Consider this: Familiarity doesn’t equal usability under pressure. A user can navigate a platform hundreds of times and still freeze the moment there’s another variable to consider in the dynamic, such as a train pulling in, a clock ticking, a decision that needs to happen now. The interface hasn’t changed but the human interacting with it has in some way. That’s exactly what controlled testing misses.
This is where continuous user testing earns its place: not as a one-time validation exercise before launch, but as an ongoing practice. If you only test once before launch, you lose sight of how your product is actually being used as it evolves.
3. If an error is possible, your user will make it
If your user can make an error, your design has already failed them. Not because users are careless, but because distracted, confused people will always find the path of least resistance. And, if that path leads somewhere wrong, that’s a design problem, not a user problem.
The goal isn’t better error messages. Rather, it’s about making errors structurally impossible. This shows up in small, elegant ways across products you probably use every day. For example:
- Gmail asks whether you forgot an attachment before you send an email that mentions one.
- Booking interfaces disable past dates rather than letting you select them and hit a wall.
- Calendly removes the entire category of scheduling errors by making the wrong choice unavailable.
For your UX research practice, this principle reframes how you read testing data. Error patterns aren’t just bugs to flag but signals pointing to deeper design assumptions worth interrogating.
“The ideal state is where the user can’t make an error,” as James puts it.
Where users consistently make the same mistake, your interface is communicating something it shouldn’t be. This is where mobile app usability testing becomes particularly valuable. Errors that go unnoticed on Desktop often surface immediately when users are navigating on a small screen, mid-task, with one hand. The conditions are closer to real life and the friction points show up faster for it.
The Metric that Actually Matters
UX metrics tend to gravitate toward what’s easy to measure: task completion rates, time on task, error frequency. While useful, they don’t always capture what good design actually achieves.
A better question to ask is: Does your product fit seamlessly into people’s lives? Do they have to think about it?
Most people don’t experience Apple Pay as a UX triumph. In fact, most people don’t know what UX design is. And that’s okay. Consumers just want to pay for things without thinking about it. That invisible quality of good design is where the achievement lies: at its best, it removes itself from the user’s awareness entirely.
That standard is worth bringing into how you frame your user research goals. The question isn’t just whether users completed the task but whether they had to think about it, whether it slowed them down, and whether they’d notice if it disappeared tomorrow.
Design for the Distracted Human in Front of You
Don’t assume your product is the most important thing in your user’s life right now. It probably ranks somewhere between a work email and what to have for dinner. They’re distracted, in a hurry, and your interface is one of many things competing for a fraction of their attention.
Remember: The most valuable user testing sessions aren’t the cleanest ones. They’re the ones that replicate the pressure, distraction, and competing demands your users bring with them every day. That’s where the real gaps show up and where the real improvements begin.
