Waiting Changes Everything

The Number You’re Optimizing Is the Wrong One

You know your load times. You track how fast your pages open, obsess over shaving seconds off, and treat speed as a design value. It is. The only problem is that users don’t experience your product the way a stopwatch measures it.

Ask users to rate a product that made them wait eight seconds and they’ll tell you it was slow and frustrating. Show them a product that also made them wait eight seconds but ended the wait with a clean, confident transition into a fully loaded state, and a meaningful share of them will tell you it felt fast. Same duration. Different memory. You didn’t change the clock. You changed the shape of the experience. This is not a perception trick or a design hack. It’s how memory actually works. And once you understand it, you’ll see why optimizing for speed alone is only half the problem.

Memory Doesn’t Record Time. It Photographs It.

Here’s the thing about how people remember experiences: they don’t play them back like a film. They reconstruct them from a handful of moments. Psychologists Barbara Fredrickson and Daniel Kahneman tested this directly, showing participants short and long versions of the same emotional film clips while measuring their real-time reactions. When asked afterward to evaluate the experience, people’s ratings had almost nothing to do with how long the clip ran. Their evaluations were “determined by a weighted average of ‘snapshots’ of the actual affective experience, as if duration did not matter.” The technical term is duration neglect. The time spent in an experience contributes far less to its remembered quality than two specific moments: the peak, the most intense point in either direction, and the end, the final moments before it closes. This is the peak-end rule, and it rewrites how you should think about every waiting state in your product. Not as a technical handoff from loading to loaded, but as an experience with a shape that users will carry away with them. Kahneman’s team made this concrete in a way that’s hard to forget. In one experiment, participants submerged a hand in 14-degree water for 60 seconds. In a second trial, they submerged the other hand for 90 seconds: the same 60 seconds of cold, then 30 more as the temperature rose slightly to 15 degrees. Still uncomfortable. Still cold. Objectively more total pain. When researchers asked participants which trial they’d prefer to repeat, a significant majority chose the longer one. They chose more pain because it ended better. The ending rewrote the memory of the whole thing (Kahneman, Fredrickson, Schreiber & Redelmeier, 1993). What this means for product design is not subtle. If users sit through a ten-second loading screen that ends by snapping them into a broken or half-ready state, they experience the worst ending your wait can offer. If that same ten-second wait ends with a smooth, complete, confident arrival, they remember the wait differently. You didn’t change the duration. You changed the snapshot that gets stored. Filled time also helps. Research on temporal cognition shows that time perceived as occupied or purposeful feels shorter than empty, formless waiting (Zakay & Block, 1997). A progress bar, an animation, a message that tells the user what’s happening: these aren’t decoration. They change how the wait feels while it’s happening. But the ending still dominates the memory. You can fill a wait beautifully and still ruin it with a bad landing. Keep that hierarchy clear.

What Uber Learned About Empty Waiting

In 2018, Uber launched Express POOL, a shared ride option with slightly longer wait times than standard uberPOOL. The product was cheaper. The routes were more efficient. On paper, the trade-off made sense. In practice, users were cancelling rides at a rate that made the economics uncomfortable. Uber’s behavioral science team investigated. The wait time itself wasn’t the killer. What was killing adoption was the experience of the wait: a screen with a static graphic and a countdown, nothing moving, nothing explaining, just time passing with no visible progress. Users had no sense of what was happening or how close they were to it being over. The wait felt endless because it was formless. The team redesigned the waiting screen with three things: a live animated map showing the driver’s actual movement, step-by-step explanations of what was happening behind the scenes as the match was being made, and a clear resolution moment when the ride was confirmed. They A/B tested the redesign against the original. The result was an 11 percent reduction in post-request cancellations (Kamat & Hogan, 2019). The wait time was identical in both conditions. What changed was the shape of the experience and what users saw at the moment it ended. This is a product team taking the science and applying it directly. Filled time, operational transparency, and a designed ending. Not a faster product. A better-remembered one.

The End Test

Here is the shift this chapter is asking for. Before you ship any flow with a wait state, a loader, a file upload, a save confirmation, a checkout process, a form submission, stop and ask one question: what is the last thing the user sees and feels at the moment the wait resolves? That moment is disproportionately what they’ll carry away. It’s the snapshot that becomes the rating, the review, the word-of-mouth, the decision about whether to come back. If it’s a spinner that disappears into an unfinished page, you’ve handed them a bad ending. If it’s a confident arrival into a state that feels complete and responsive, you’ve handed them a good one. Call this the End Test. It doesn’t replace performance optimization. Faster is still better. But it sits alongside speed as an equal discipline, one that most teams never apply. Run it on every wait state in your product. Check what the user lands on. Check whether it feels resolved or abandoned. Check whether the final moment of the wait communicates that something good just happened. Most loading screens fail this test not because the load is slow but because no one thought about what success looks like from the user’s side. The screen disappears and the product appears, and that transition gets no design attention at all. It’s treated as a technical event rather than an experiential one. It is not a technical event. It is the moment judgment forms.

The Spinner Is Not Neutral

You’ve probably spent hours designing the state users see when your product is working well. You’ve spent almost no time designing the state they see while they wait for it. That’s the distortion this chapter is naming. Duration is measurable, so it gets measured. The shape of a wait is harder to quantify, so it doesn’t make it into the sprint. But what users remember, and what determines their overall feeling about your product, is not the clock time. It’s the peak and the end. The snapshot, not the film. The spinner is the last thing users see before they decide what they think of you. Design it like that’s what it is.