[thelist] maximum page load for users leave?

aardvark roselli at earthlink.net
Wed Dec 4 21:17:01 CST 2002

> From: "Liam Delahunty" <ldelahunty at britstream.com>
> >by the way, i don't buy the "under 10 seconds" argument
> I do. Sometimes. In (real-life real people), user tests I performed,
> I've found that if the page takes a while to show *anything* (nested
> and hits back or refresh.

my own user tests suggest 8 seconds is the max... interestingly, that
number is the same as another number that was found when testing
operating systems -- how long will a user wait before hitting the
three-finger salute?  8 seconds... how to change that?  make icons
animate while they wait (etc.)...

> On the other hand if the page has been crafted so _something_ starts
> to load and the user is aware that more is to come then they'll wait

progressive table layouts, for one, can help... stack your tables...
so, if you are properly re-using images, your header/navbar can
perhaps display as soon as the code comes down for it, while it
finishes downloading the rest of the content... this can make a >8sec
load time feel like a lot less because of the perception...

> >not everybody's on cable, and people who aren't have more patience
> That's true, but so is the converse, (A)DSL, cable, etc has made users
> even hungrier and less patient. A sites problem _could_ be impatient
> cable users who are looking for a broadband fix promised by the
> telecom adverts that the Internet just can't deliver to regular users

and this is why friends refuse to surf on my 56k line... even the odd
4sec page makes them all grumpypants...

i've seen that effect a lot when testing...

Read the evolt.org case study
Usability: The Site Speaks for Itself
ISBN: 1904151035

More information about the thelist mailing list