Almost every product team I have worked with measures the same things: time to first action, conversion rate through the funnel, retention by cohort. These metrics are useful, but they describe symptoms. The underlying variable they are all tracking, indirectly, is friction. Not the friction users complain about — that one gets fixed in a sprint. The friction users do not complain about, because they have already left.
Two kinds of friction dominate the experiences I see breaking down. They are usually invisible to the people building the product, because the people building it have stopped feeling them. Both are the territory where the strongest UX work actually happens.
Cognitive friction
Cognitive friction is the load the product puts on the user's attention, working memory, and decision-making. Every additional choice, every unfamiliar pattern, every screen that asks the user to remember something they were told two steps ago adds to it. Most products accumulate cognitive load gradually, one well-intentioned feature at a time, until the experience becomes harder to use than it should be for what it does.
The symptoms are quiet. Users hesitate before tapping. They skip sections they intended to engage with. They open the app, look at the home screen, and close it without doing anything. The product is not broken; it is exhausting. Exhausting products do not get used.
The discipline of reducing cognitive friction is mostly subtraction. Fewer options at any single decision point. A clearer hierarchy of what matters. Consistency of interaction patterns so the user does not have to relearn every screen. Defaults that make sense for most people, with the option to deviate hidden until needed. Saying no to features that add load without adding proportional value.
Emotional friction
Emotional friction is the resistance the user feels because of how the product makes them feel, separate from what it makes them do. A goal-tracking app that quietly communicates "you're behind" every time the user opens it generates emotional friction. So does an onboarding flow that demands the user be vulnerable before they've earned any trust. So does a notification that interrupts at the wrong moment, in the wrong tone.
Most emotional friction is not malicious. It is the result of teams designing screens in isolation, without thinking through what the screen will feel like in the context of the user's actual day. The push notification that reads as helpful on a Wednesday afternoon reads as nagging at 7 AM on a Saturday. The streak counter that motivates on day five demotivates on day twenty-three when the user breaks it.
The most sophisticated UX teams are the ones who understand that the user's emotional state is part of the product experience — not a marketing concern, but a design variable.
What the work looks like
Reducing both kinds of friction requires the team to do a few unglamorous things consistently.
- Map the experience as a sequence in time, not as a set of screens. Where does the user get tired? Where do they get discouraged? Where are they asked to do too much in one step?
- Audit defaults aggressively. Every default the product makes for the user is a friction reduction. Every required choice is a friction increase. The ratio matters.
- Match tone to moment. The product should know whether this is a momentum moment, a recovery moment, or a quiet check-in — and respond accordingly. This is harder than copywriting a single voice and applying it everywhere.
- Treat consistency as a feature. Users who can predict what will happen when they tap something are users whose cognitive load is low. Novelty in interaction patterns is almost always a net loss.
- Design for the user who is tired. Most product testing is done by people who are alert and interested. Most actual product use happens by people who are tired, distracted, and operating at half capacity. Designing for the alert user produces products that fail for the tired one.
The flow-state target
The opposite of friction is not speed. It is flow. A product in flow does not feel fast; it feels effortless. The user's attention stays on what they are trying to accomplish, not on the product itself. Decisions are minimal. The next action is obvious. The experience does not interrupt the user's thinking; it supports it.
Flow is harder to design for than features, which is why most teams retreat to feature work. But it is the variable that separates products that get used from products that get downloaded. Users do not consciously articulate "this product has low cognitive and emotional friction." They say "this just works," or more often, they say nothing and just keep using it. That second outcome is the design target.