So from the user’s perspective, interruptions while waiting for a particular page to load don’t represent a natural interruption: users haven’t yet achieved their goal, which may make them less tolerant of delays, meaning that pages need to load quickly so that the user’s experience can proceed smoothly. How fast is really fast? This is a meaningless question because there is no single answer for three main reasons. First, the answer depends on what outcome you consider, such as abandonment, user satisfaction, or task performance.
Different studies focus on different perspectives and produce different results. Second, if you plot how many users stay on a site as a function of the number of delays they experience, you won’t see a clear step luxembourg mobile database from 100% to 0% after X seconds. Instead, you see a smooth distribution that looks like this: The impact of page speed on user retentionSo we have to ask: which point on this curve are we aiming for?
In other words, how much are we investing in speed on the one hand, and how many users will we lose on the other? In the end, the impact of latency varies by context and situation. News sites, shopping sites, and travel sites are often part of different types of user experience processes, and the entire curve above may look different for each one. Even within a specific environment, website design and user behavior can change over time.
Although this is more difficult than we would like, distributing the results across different performance levels still contains useful hints. In particular, this distribution reveals how many users we might lose (or are losing) at a given level of performance.