The 10 UX problems we see in almost every health tech product
These aren't edge cases. They show up in well-funded products with strong engineering teams. They're not bugs — they're design decisions that were never revisited.
We'll review 3 core flows and share specific findings within 5 business days.
How this was built
20+ health tech products designed and audited since 2018. These are the patterns that kept appearing.
Clinical AI, telehealth, patient apps, digital therapeutics, medtech — early-stage startups and established platforms alike. The problems in this report aren't theoretical. We encountered them while shipping products that are now live.
Onboarding asks too much, too soon
Common pattern
Suggested improvement
Why it happens
Teams confuse what the system needs with what users should provide upfront. The result is a registration flow that feels like a medical intake form before the user has seen any value.
What it costs
High abandonment during signup. Users who complete it arrive frustrated, not curious. In health tech, first impressions are often irreversible — a bad start colours the entire relationship with your product.
What good looks like
Lead with a single moment of value. Collect information progressively — as it becomes relevant, not all at once. The user should experience what your product does before being asked to invest in setup.
“If your onboarding feels like paperwork, you've lost the user before they've started.”
Two audiences, one interface
Common pattern
Suggested improvement
Why it happens
Health platforms often serve both patients and providers. Building separate experiences feels expensive, so teams compromise with a single interface that tries to serve everyone.
What it costs
Patients find the interface clinical and intimidating. Providers find it oversimplified. Neither group feels the product was built for them — and both have lower tolerance for friction than you expect.
What good looks like
Identify where the experiences need to diverge and design them separately. Shared infrastructure is fine. Shared interfaces are where things break down. Even small differences in tone and density make a significant impact.
“When you design for everyone, you design for no one.”
Data overload on clinical dashboards
Common pattern
Suggested improvement
Why it happens
Clinical products surface everything because everything could matter. Over time, dashboards accumulate data points without anyone asking what actually needs attention right now.
What it costs
Critical alerts get buried next to routine information at the same visual weight. In clinical settings this isn't just a usability issue — it's a patient safety issue. Missed alerts have real consequences.
What good looks like
Build clear information hierarchy. Lead with what requires action, separate it visually from background data, and let providers drill into detail only when they choose to. Less on screen doesn't mean less capability.
“When everything is important, nothing is.”
No empty states or first-time guidance
Common pattern
Suggested improvement
Why it happens
Teams test with sample data and never experience what a brand new user sees — which is usually nothing. Blank screens, empty tables, zero context.
What it costs
New users don't understand what the product does or where to start. This is especially damaging in health tech where users may already feel anxious. A blank screen reads as broken.
What good looks like
Treat empty states as a feature, not an edge case. Welcome users, explain what will appear here and why, and offer a clear first action. Your empty state is your first real conversation with the user.
“Your empty state is your product's first impression. Make it count.”
Medical terminology in patient-facing flows
Common pattern
Suggested improvement
Why it happens
Health products are built by people comfortable with clinical language. That language leaks into patient-facing interfaces because nobody flags it during development.
What it costs
Patients feel confused, intimidated, or excluded. They make errors on forms they don't understand. Features get avoided. This creates real downstream problems in care delivery.
What good looks like
Write the way a good doctor actually talks to a patient. 'What brings you in today?' does the same job as 'Indicate your chief complaint.' Plain language isn't about dumbing things down — it's about removing unnecessary barriers.
“Plain language isn't less precise. It's more accessible.”
Recognising any of this in your product?
We'll review 3 core flows and share specific findings within 5 business days — for free.
Inconsistent UI across features
Common pattern
Suggested improvement
Why it happens
Products grow feature by feature, often with different teams or contractors. Without a shared design system, every new feature introduces its own patterns, spacing, and interaction logic.
What it costs
Users don't consciously notice inconsistency — but they feel it. It erodes trust in ways that are hard to trace. It also slows development over time as every new feature starts from scratch.
What good looks like
Invest in a shared component library early. It doesn't need to be elaborate. Even a small set of shared patterns for buttons, forms, and navigation creates consistency that compounds over time.
“A design system isn't overhead. It's the foundation that lets you move faster.”
Poor error handling in high-stakes flows
Common pattern
Suggested improvement
Why it happens
Error states are the last thing teams design and the first thing cut when timelines get tight. Generic messages get shipped because they're easy to implement.
What it costs
In healthcare, a vague error creates real anxiety. Did my prescription go through? Did my results upload? When users can't answer these questions, trust breaks — and they may take actions that create clinical risk.
What good looks like
Every error in a health product should answer three questions: what happened, what it means for the user, and what they should do next. Offer a direct path to resolution whenever possible.
“Vague errors are never acceptable when the stakes are clinical.”
Mobile app feels like a shrunken web app
Common pattern
Suggested improvement
Why it happens
Most health tech products start as web apps. When mobile ships, it's often a responsive version of the desktop experience — not something designed for mobile from the ground up.
What it costs
Tap targets are too small. Navigation borrows desktop patterns. Swipe gestures are missing. The app fights the phone instead of working with it. Users notice immediately, even if they can't articulate why.
What good looks like
Mobile is not a smaller screen — it's a different context. Design for thumb zones, platform conventions, and the reality that your user might be standing in a waiting room with one hand free.
“If it feels like a website in an app wrapper, your users will notice.”
Privacy settings are buried or confusing
Common pattern
Suggested improvement
Why it happens
Privacy controls are complex to build, so they get simplified or hidden. Health apps collect deeply sensitive data but consistently treat permission management as an afterthought.
What it costs
Users who care about their health data — most of them — lose trust quickly when they can't find or understand privacy controls. In health tech, trust isn't optional. It's the entire foundation of retention.
What good looks like
Make privacy controls visible, understandable, and easy to act on. Surface key controls during onboarding. If they're hard to find, users assume you're hiding something.
“In health tech, trust isn't a feature. It's the foundation.”
The demo looks great. The real product doesn't.
Common pattern
Suggested improvement
Why it happens
Teams invest heavily in making demos compelling for investors and sales calls. They use curated data, happy paths, and carefully chosen screen sizes. The real product lives in a messier world.
What it costs
The gap between demo and product erodes credibility with everyone who matters: users who feel misled, buyers who feel oversold, and the team itself, which starts to lose confidence in what they've built.
What good looks like
Test with real data early and often. Use realistic name lengths, edge case content, and actual user scenarios. If your product only looks good with sample data, it doesn't look good yet.
“The gap between your demo and your product is the most honest measure of your design maturity.”
Self-assessment: how does your product score?
Tick the statements that are already true for your product today.
Step 1 of 4
First Contact
Assessment results
Start the checklist
Select the statements that already apply to your product.
UX Audit
If any of this looks familiar, let's talk.
We'll review 3 core flows and share specific, prioritised findings within 5 business days — no strings attached.