Does the size of website matter and when?

[TomLessing]

What are the arguments for keeping the download size of website as
small as possible and what are the arguments against?

In my experience a very hard topic, business almost always wants the
nice funky look, the one that looks like it cost a million to design
but uses huge amounts of bandwidth.

[AJK]

Jakob Nielson (www.useit.com) will talk your ear off for a whole day,
if not more, regarding download size of websites. :)

A lot of people want a kewl looking site and some people just don't
care; they just want the info. I can only speak for myself as I multi-
task, which means that I browse or read another site, while the large
one open in the background, but I have a limit. If the site takes too
long I close the window. If there is another option for the info I am
looking for, then I will try the alternative and that is why small
pages are important.

Now what researches like Nielson don't usually talk about are the
exceptions. Depending on what kind of information you have (are you
unique in your field?), people might have more patience for a larger
download size. If I am a member of the 5FM site, I will have more
patience. If I want to see photos taken by you and it is only
available on your site, then I will have more patience. If I am just
looking for info that I can find on numerous sites, then my patience
will be less.

So, depending on the uniqueness (rare content, specialised content,
club or organisation) of your content, people might have more
patience. My suggestion is always in these situations when you want to
have a nice look: Load the content which people can read, first. It
gives the user something to read while the rest of the design loads.

[Shaun O'Connell]

Hi Tom,

Benefits of optimising website output sizes:

1. The server saves on bandwidth costs. Especially relevant here in SA, if the server is hosted in SA.
2. The client saves on bandwidth costs. Especially relevant here in SA.  Bandwidth is expensive in SA.
3. The client saves on download time.  Clients typically have a fraction of the bandwidth the server has.  Yes, the server will send pages quicker if there is less to send, but this is hardly noticeable.
4. The User might be less inclined to navigate away if the website seems more responsive.  (See AJK's points)

Drawbacks of optimising website output sizes:

1. Few developers in SA can do this effectively.  It is improving though.
   
2. Boring?  (Not necessarily - See below)

This is where Web Standards based web development starts to shine.

The gist of which is:

• Optimise the un-cacheable content (HTML) by reducing its byte overhead. (Semantic, presentation-less HTML)
• Use CSS to serve as your style repository, defining layouts, adding decorative images.
• Let the client's browser cache all the CSS and images.  Better responsiveness the second time around.
  
• "Progressively Enhance" the User's interaction through JavaScript.

You don't need expensive overheads like Flash to make a site look like a million-bucks worth.  Check out http://cssvault.com/ and http://www.cssbeauty.com/ for a few examples.
No more need for "Printer Friendly Pages", or serving alternative content to different browsers or platforms.  Just serve different CSS.

If you've got any further questions regarding web standards, don't hesitate to ask.

Cheers,
Shaun

[Helga]

Completely agree with Shaun on this, recently there was a study done
in the UK on how long customers are prepared to wait for a loading
homepage. I can't remember the exact detail but 1/6 of a second was
seen as instantaneous loading, 1-5 seconds were seen as reasonable
time for a homepage load and 5-10 seconds customers had an
overwhelming urge to leave and go elsewhere. So pageload times is a
very important competitve lead for business. It is so important for e-
commerce sites, that I once had to work on a re-design project to
bring the homepage size down to make the site more competitive, the
competitor had his homepage at 120K and we were at 150K we had to re-
develop the whole page to come in at 80K only for the competitor to
launch their page at 70K 4 months later, we therefore had a regular
yearly enhancements to bring our homepage size down.

From my experience 99% of customers are task-based and they are
unequivical in their need for fast clear sites, however much the
business sees their site as a online-broshure where they can create
this nice interactive brand experience only designers browse. If the
company is into making money they need to see the web in a whole new
light focussing on customer experience and giving their clients what
they are seeking and generally it is not a brand experience but rather
a research facility online.

On May 6, 3:29 pm, "Shaun O'Connell" <ndor...@gmail.com> wrote:
> Hi Tom,
>
> Benefits of optimising website output sizes:
>

>    1. The server saves on bandwidth costs. Especially relevant here in

>    SA, if the server is hosted in SA.

>    2. The client saves on bandwidth costs. Especially relevant here in

>    SA.  Bandwidth is expensive in SA.

>    3. The client saves on download time.  Clients typically have a

>    fraction of the bandwidth the server has.  Yes, the server will send pages
>    quicker if there is less to send, but this is hardly noticeable.

>    4. The User might be less inclined to navigate away if the website

>    seems more responsive.  (See AJK's points)
>
> Drawbacks of optimising website output sizes:
>

>    1. Few developers in SA can do this effectively.  It is improving
>    though.
>    2. Boring?  (Not necessarily - See below)

>
> This is where Web Standards based web development starts to shine.
>
> The gist of which is:
>

>    - Optimise the un-cacheable content (HTML) by reducing its byte
>    overhead. (Semantic, presentation-less HTML)
>    - Use CSS to serve as your style repository, defining layouts, adding
>    decorative images.
>    - Let the client's browser cache all the CSS and images.  Better

>    responsiveness the second time around.

>    - "Progressively Enhance" the User's interaction through JavaScript.

>
> You don't need expensive overheads like Flash to make a site look like a

> million-bucks worth.  Check outhttp://cssvault.com/andhttp://www.cssbeauty.com/for a few examples.

> No more need for "Printer Friendly Pages", or serving alternative content to
> different browsers or platforms.  Just serve different CSS.
>
> If you've got any further questions regarding web standards, don't hesitate
> to ask.
>
> Cheers,
> Shaun
>

> On Tue, May 6, 2008 at 1:22 PM, TomLessing <tom.lessing.lsqu...@gmail.com>

> wrote:
>
>
>
>
>
> > What are the arguments for keeping the download size of website as
> > small as possible and what are the arguments against?
>
> > In my experience a very hard topic, business almost always wants the
> > nice funky look, the one that looks like it cost a million to design

> > but uses huge amounts of bandwidth.- Hide quoted text -
>
> - Show quoted text -

[TomLessing]

Does anybody have some solid statistics on the topic that is available
online?

[AJK]

http://www.useit.com/alertbox/9703a.html
http://www.useit.com/alertbox/990530.html

Response Times: The Three Important Limits
http://www.useit.com/papers/responsetime.html

[phi...@gmail.com]

Here's a different take from Jared Spool. They did research about
PERCEPTION of download time vs ACTUAL download time.


http://www.uie.com/articles/download_time/


"If people can't find what they want on a site, they will regard the
site as a waste of time (and slow). But, when users successfully
complete tasks on a site, they will perceive their time there as
having been well spent.

Jakob Nielsen tells designers to focus efforts on improving actual
page download times on their sites. But what we're seeing leads us to
wonder if it's worth the resources to make web pages load like
lightning. Instead, we're wondering: When users are complaining about
the download speed of your site, what are they actually complaining
about? Are you better off making the site load faster or ensuring that
users complete their tasks?"


--phil--

On May 6, 5:06 pm, AJK <ajk...@gmail.com> wrote:
> http://www.useit.com/alertbox/9703a.htmlhttp://www.useit.com/alertbox/990530.html

[AJK]

"Are you better off making the site load faster or ensuring that users
complete their tasks?"

Are they actually competing aspects or should you actually try to
improve both?

[Helga]

They really go hand-in-hand, they are both very important, the fastest
site will not be good if users can't actually complete a task (I find
nielsens site often irrates me in this way, so fast yet ugly as hell I
am bored at even looking for stuff and then getting the 2000 research
paper enough to drive you mental).

So both is a given and should be considered, but make sure you inform
customers of the site's purpose right in the beginning as you have 4
seconds aprox.(According to a HCI research paper I will track down for
you) to inform customers of the sites purpose before they leave to
find their product elsewhere.