There are three important response time limits in user interface design, and this has remained constant since 1968, says usability guru Jakob Nielsen. Those three time limits?
- 0.1 second is about the limit for having the user feel that the system is reacting instantaneously […] for users feeling that they are directly manipulating objects […] as opposed to feeling that they are ordering the computer to manipulate objects for them.
- 1.0 second is about the limit for the user’s flow of thought to stay uninterrupted. […] The limit for users feeling that they are freely navigating the command space without having to unduly wait for the computer. […] Users notice the delay and thus feel the computer is “working” on the command.
- 10 seconds is about the limit for keeping the user’s attention focused on the task. For longer delays, users will want to perform other tasks while waiting for the computer to finish. […] More than 10 seconds, and you break the flow.
It’s worth also looking at Nielsen’s Powers of 10, detailing further time scales of user interaction. My summary:
- 1 minute is the limit in which users should be able to complete simple tasks.
- 10 minutes is a long visit to a website.
- 1 hour is the limit for completing most web-based tasks.
- 1 day is the maximum turnaround for (good) customer service and the start of habitual routines.
- 1 week is another common time frame for habitual routines or complex tasks requiring extensive research.
- 1 month is the time it might take for business processes, as various people need to be involved in decisions.
- 1 year is the time it takes for organisational changes to start taking place, and is how long it takes to nurture experienced users (for whom interfaces can be more complex).
- 10 years is how long it takes deep expertise to develop for a complex system.
- 100 years is sufficient for complex social changes to take place.