Progress bars don't have to be boring (ask Coolblue's little truck)

• Goal-gradient effect. The closer to done, the stronger the pull. Hull's rats ran faster as they neared the food box, and people buy coffee faster as the stamp card fills up. The last stretch of your bar is motivational rocket fuel.
• The Zeigarnik effect. Unfinished tasks occupy the mind rent-free. A visibly half-full bar creates an open loop your user itches to close.
• Endowed progress. People work harder toward a goal they feel they've already started, which makes the starting point of your bar a design decision.
• The peak-end rule. People judge an experience by its most intense moment and its ending, not its average. A small celebration at 100% retroactively upgrades the whole wait.

The evidence: the goal-gradient effect goes back to Clark Hull's maze experiments in the 1930s and was replicated in humans decades later with coffee stamp cards. Streaks and levels borrow the same psychology. A progress bar is just the smallest, cheapest place to apply it.

That's the whole trick Coolblue pulled off. Nobody rewatches a percentage. A truck driving to your house is a small story though, and a story has an ending you want to see. The order status went from something you check to something you show to whoever happens to sit next to you.

• Match the metaphor to the promise. Coolblue's truck works because the bar is about a delivery. A setup wizard can be a building going up, a coffee order can be a cup filling. Random confetti on a database sync is just decoration.
• Never move backwards. A bar that jumps backwards breaks trust in one frame. Estimate conservatively, so that when your numbers are off the bar skips ahead instead of retreating.
• Land the ending. Attention peaks in the final stretch, so that's where the craft belongs. A doorbell at 100% is worth more than any amount of polish at 40%.
• Stay honest. Fun is not a license to fake. If the bar says 90%, the work should be about 90% done. The first time users catch a lying bar, every future wait feels longer.
• Design for the hundredth viewing. A checkout someone sees once can be theatrical. A sync bar they see daily should be calm. The more often someone sees an animation, the quieter it needs to be.

Loyalty cards proved it decades ago: in Nunes and Drèze's car wash study, a 10-stamp card with 2 stamps pre-filled was completed almost twice as often as an empty 8-stamp card, identical effort. The same head start works on anything with a bar. Start the profile meter at 20% for filling in an email address, or open the order tracker with the first stage already ticked.

The principle: where a bar starts is a design decision, not a data readout. Give people a running start and more of them reach the end.