Pagination vs. Scrolling

I just read a pre-publish paper by a colleague that discusses this
exact issue. (you should see it online soon) There is some evidence
that even late adopters of the web have no difficulty and in fact
expect to scroll down. That has also been my gut sense for quite a
while as well. One of the difficulties comes in a logical place to
end the page... particularly with text. Ideally, any pagination
system should play well with the browsers print function. This can be
a problem.

Mark

On Jul 17, 2007, at 6:23 PM, Wilson, Russell wrote:

> Have there been any studies on when it is most appropriate to
> switch to
> paging a
>
> list versus scrolling?
>
>
>
> In other words, if I have a list of 20 items, I might use a simple
> scrolling mechanism
>
> to move through the list. But if the list grows to 2000, I might
> use a
> paging mechanism
>
> (with 20 item windows).
>
>
>
> What number of items (if any) is the breaking point to make the
> switch?
> Can anyone
>
> share some common practices?
>
>
>
> Best regards,
>
> Russ
>
>
>
>
>
> Russell Wilson
>
> Director of Product Design, NetQoS
>
> www.dexodesign.com
>
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://beta.ixda.org/guidelines
> List Help .................. http://beta.ixda.org/help
> Unsubscribe ................ http://beta.ixda.org/unsubscribe
> Questions .................. list at ixda.org
> Home ....................... http://beta.ixda.org

Russ,

This is one of those questions that begs for context. I don't think there is
a good answer in the abstract -- it depends significantly on the task, which
in turn cannot be divorced from the overall task flow, previously used user
conventions in the rest of the site/application, target access speed, server
speed, user types, etc.

We discuss the pro's and con's of this question with nearly every client
we've worked with.

Some pro's of pagination:

-- If the number of lines per page stays "above the fold" then a user can
get through a small to medium size list without having to move their mouse
at all (click, next, next, next).
-- If the number of lines per page stays "above the fold" then a user never
loses sight of other navigation elements in the banner/navigation areas.

Some con's of pagination

-- Users with fast connections and who are fast at moving around in a
site/application are frustrated with having to wait for every screen
refresh. Letting users set the number of items shown in each page mitigates
this frustration, but works against some of the pros' above.
-- Lists with hundreds (to thousands) of items are really frustrating for
just about anyone to paginate through.

Some pro's of scrolling

-- Assuming a good connection and a good server response, it can be the
fastest way to get to a particular line item.
-- Going "backward" through the list is nearly effortless.

Some con's of scrolling

-- Slow server response can be very frustrating. A long list that takes 20
seconds (and upward) to display will not be popular.
-- Regular users can get eye strain and fatigue scrolling though huge lists.

Other thoughts:

We always try to eliminate really long lists by giving users filters, sorts,
categories, quick search capabilities, etc. so that they neither have to
scroll or paginate through 100's to thousands of line items. In a recently
completed project we did with Cisco (their application for processing and
creating service contracts on all their products) a single service contract
for a major customer like GE could run to 50,000 line items. It was critical
to give users intuitive controls for pairing lists down to manageable size
for viewing.

We have yet to consistently come down on one or the other approach. One or
the other approach never satisfies everybody.

Regards,
Joseph

Joseph Selbie
Founder and CEO
Tristream
530-477-5777 x-202
530-277-6326 cell
jselbie at tristream.com
www.tristream.com

-----Original Message-----
From: discuss-bounces at lists.interactiondesigners.com
[mailto:discuss-bounces at lists.interactiondesigners.com] On Behalf Of Wilson,
Russell
Sent: Tuesday, July 17, 2007 3:23 PM
To: IXDA list
Subject: [IxDA Discuss] Pagination vs. Scrolling

Have there been any studies on when it is most appropriate to switch to
paging a

list versus scrolling?

In other words, if I have a list of 20 items, I might use a simple
scrolling mechanism

to move through the list. But if the list grows to 2000, I might use a
paging mechanism

(with 20 item windows).

What number of items (if any) is the breaking point to make the switch?
Can anyone

share some common practices?

Best regards,

Russ

Russell Wilson

Director of Product Design, NetQoS

www.dexodesign.com

________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... discuss at ixda.org
List Guidelines ............ http://beta.ixda.org/guidelines
List Help .................. http://beta.ixda.org/help
Unsubscribe ................ http://beta.ixda.org/unsubscribe
Questions .................. list at ixda.org
Home ....................... http://beta.ixda.org

> Have there been any studies on when it is most appropriate to switch to
> paging a list versus scrolling?

There are threads on this as well:

http://beta.ixda.org/search.php?tag=pagination

-r-

What about splitting long articles into few pages?

Pro's:
User always can create a bookmark on the page that he had to stop to read

Con's:
Someone who is going to read all article had to use pagination.

--
Maxim
http://maxim.deast.info/

I don't buy into the conspiracy thing here, if only in name.

Monetization is a realistic and ethical design constraint. Designers need to understand how things (salaries and other resources) get paid for.

For long lists, pagination makes perfect sense partly because it is so easy - and prints fairly well. User options for number of items to display are great.

Mark

On Wednesday, July 18, 2007, at 08:12AM, "Kevin Doyle" <kbdoyle at gmail.com> wrote:
>Another Con:
>It's not possible to print out an article that's spread out over several
>pages unless there's an option to do so.
>
>I'm with Rob Brown on this one: pagination is a Marketing tool. I've found
>pagination only to be useful on massive information returns like record and
>internet searches.
>
>
>On 7/18/07, Maxim Soloviev <maxim at deast.info> wrote:
>>
>> What about splitting long articles into few pages?
>>
>> Pro's:
>> User always can create a bookmark on the page that he had to stop to read
>>
>> Con's:
>> Someone who is going to read all article had to use pagination.
>>
>> --
>> Maxim
>> http://maxim.deast.info/
>> ________________________________________________________________
>> Welcome to the Interaction Design Association (IxDA)!
>> To post to this list ....... discuss at ixda.org
>> List Guidelines ............ http://beta.ixda.org/guidelines
>> List Help .................. http://beta.ixda.org/help
>> Unsubscribe ................ http://beta.ixda.org/unsubscribe
>> Questions .................. list at ixda.org
>> Home ....................... http://beta.ixda.org
>>
>________________________________________________________________
>Welcome to the Interaction Design Association (IxDA)!
>To post to this list ....... discuss at ixda.org
>List Guidelines ............ http://beta.ixda.org/guidelines
>List Help .................. http://beta.ixda.org/help
>Unsubscribe ................ http://beta.ixda.org/unsubscribe
>Questions .................. list at ixda.org
>Home ....................... http://beta.ixda.org
>
>

The book Prioritizing Web Usability by Jakob Nielsen and Hoa Loranger
(2006) has a discussion on pg 45 about their test results on users
and scrolling. Mind you it was website specific which appears to mean
pages of content vs. lists but still interesting results to share with
your stakeholders.

"Obviously all users saw the first screenful(the one above the
fold). But viewing frequency dropped off rapidly after that. More
than half the users didn't scroll at all, so only 42% of users saw
any information on the second screenful........
only the most persistent 1 percent of users viewed more than seven
screens worth of information"

They give no recommendations for scrolling vs pagination, other than
make sure the first page displays enough important information that
makes the user want to investigate further.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://beta.ixda.org/discuss?post=18318

I agree with Joseph and others that context is key in determining
whether to page or scroll.

In our web portal, we've found through studies that users are
generally *willing* to scroll to get to content if they already know
it is there, and if the most important content is above the fold.
That is, many folks have become used to sloppy content writers who
can't fit everything in one page, and as long as they don't have to
scroll for the most clicked items and only scroll when they have to,
they're ok.

This breaks (again, on our portal) when content owners put content
below the fold to "lure" people to learn something. We call
them "Ads", content owners call them "Critical Content". Either way,
users are very bad at "discovering" anything below a page's fold,
because they only go there to do things they need to do, not "browse".

The whole argument is different if people need to work with records in
a tabular display. A well-designed paging system that lets people
quickly narrow long lists, but also manage large data sets, is usually
a good answer to people who *work with* many records at a time.
Unfortunately, there's usually only time to design for one audience.
Those who only need one record usually have to scroll through many
pages to find the one they want. Again, good design and filters can
alleviate that.

As an example, we have a CMS "gadget" in our portal that displays all
of the content records you can edit. Administrators need to be able
to edit any content item, so my version of this portlet has 25,000
alphabetized items in it, with no filters or "jump to page" controls.
It's horrific and forces me to circumvent the entire thing to help
people edit their content.

Bryan

----- Original Message -----
From: Mark Pawson <mark.pawson at ihs.com>
Date: Wednesday, July 18, 2007 12:02 pm
Subject: Re: [IxDA Discuss] Pagination vs. Scrolling
To: discuss at lists.interactiondesigners.com

> The book Prioritizing Web Usability by Jakob Nielsen and Hoa Loranger
> (2006) has a discussion on pg 45 about their test results on users
> and scrolling. Mind you it was website specific which appears to mean
> pages of content vs. lists but still interesting results to share
with
> your stakeholders.
>
> "Obviously all users saw the first screenful(the one above the
> fold). But viewing frequency dropped off rapidly after that. More
> than half the users didn't scroll at all, so only 42% of users saw
> any information on the second screenful........
> only the most persistent 1 percent of users viewed more than seven
> screens worth of information"
>
> They give no recommendations for scrolling vs pagination, other than
> make sure the first page displays enough important information that
> makes the user want to investigate further.
>
>
> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
> Posted from the new ixda.org
> http://beta.ixda.org/discuss?post=18318
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://beta.ixda.org/guidelines
> List Help .................. http://beta.ixda.org/help
> Unsubscribe ................ http://beta.ixda.org/unsubscribe
> Questions .................. list at ixda.org
> Home ....................... http://beta.ixda.org
>

I recently asked the owners of our LDAP admin tool to use such a
malleable approach. When they show members of a group, they provide 5
members per page with no filters. They said performance was the
reason, but I asked them to increase the pages to 20 (how bad could it
get?, it's an LDAP directory). Further, I asked them to do a "record
count" before fetching the list. If there are fewer than 30 or so
members in the group, just put them all in the page. Otherwise,
provide the 20 item pages, and add some filters.

Bryan

----- Original Message -----
From: "Martin J. Lambert" <MLambert at NBME.org>
Date: Wednesday, July 18, 2007 2:13 pm
Subject: Re: [IxDA Discuss] Pagination vs. Scrolling
To: IXDA list <discuss at ixda.org>

> > I was also wondering what the "magical number" might be for
> > switch to pagination...
> > Can I really give a user a scrollable list of 50,000 items??
> > (disregarding performance for now, which plays a large part
> > in the ultimate decision as well)
>
>
> Not a direct answer, more of a tangent: something I've been toying
> with on our intranet (though it's far from perfect yet) is to make
> this "magic number" somewhat malleable.
>
> In particular, it's very annoying to click through to the last page
> of a result set to find only a tiny number of results there. Thus,
> if my "magic number" is 100 results per page, I might actually show
> 95 or 108, as long as it brings the last page into a comparable
> range with the others.
>
> Of course, once you get past a certain point (another magic number-
> 15 pages for example), then it hardly matters because virtually no
> one will ever see that last page.
>
> As for what the magic number is, I agree with others that it all
> depends on context. I've used a similar technique with the number
> of thumbnails per row on a page full of images, where the magic
> number is based on the width of the browser window. If the last
> row only has one or two thumbnails, I'll add or subtract a little
> padding to even things out.
>
> Martin Lambert
> mlambert at nbme.org
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://beta.ixda.org/guidelines
> List Help .................. http://beta.ixda.org/help
> Unsubscribe ................ http://beta.ixda.org/unsubscribe
> Questions .................. list at ixda.org
> Home ....................... http://beta.ixda.org
>

Hi Kevin,

Totally get that, and I play the consumer/user advocate role arguing against ads nearly every day. But there is nearly always a business influence and the monetization is likely equally important if not more (to those making the final decision). My real point being that rather than have this decision forced on us, maybe we need to stop moping around and wining about it and accept it as a design constraint and create a solution that anticipates it.

Mark

On Wednesday, July 18, 2007, at 03:22PM, "Kevin Doyle" <kbdoyle at gmail.com> wrote:
>I think that a pleasant user experience is good Marketing and supersedes any
>"monetization" constraints. How many pages will a user click through if they
>have to worry about a whole bunch of images, ads and Flash movies every time
>they click "next". And pagination doesn't print well at all -- you have to
>go to each page and print out things separately. You might be able to have a
>"print-friendly version" of the article available, but I bet that once the
>user clicks on the ad-free print-friendly, full article version, they're
>going to finish reading the article there. Besides, with some good CSS
>development, you don't even need to have a print-friendly verison of the
>article available -- you just create a style sheet just for printing.
>
>On 7/18/07, Mark Schraad <mschraad at mac.com> wrote:
>>
>> I don't buy into the conspiracy thing here, if only in name.
>>
>> Monetization is a realistic and ethical design constraint. Designers need
>> to understand how things (salaries and other resources) get paid for.
>>
>> For long lists, pagination makes perfect sense partly because it is so
>> easy - and prints fairly well. User options for number of items to display
>> are great.
>>
>> Mark
>>
>>
>> On Wednesday, July 18, 2007, at 08:12AM, "Kevin Doyle" < kbdoyle at gmail.com>
>> wrote:
>> >Another Con:
>> >It's not possible to print out an article that's spread out over several
>> >pages unless there's an option to do so.
>> >
>> >I'm with Rob Brown on this one: pagination is a Marketing tool. I've
>> found
>> >pagination only to be useful on massive information returns like record
>> and
>> >internet searches.
>> >
>> >
>> >On 7/18/07, Maxim Soloviev <maxim at deast.info> wrote:
>> >>
>> >> What about splitting long articles into few pages?
>> >>
>> >> Pro's:
>> >> User always can create a bookmark on the page that he had to stop to
>> read
>> >>
>> >> Con's:
>> >> Someone who is going to read all article had to use pagination.
>> >>
>> >> --
>> >> Maxim
>> >> http://maxim.deast.info/
>> >> ________________________________________________________________
>> >> Welcome to the Interaction Design Association (IxDA)!
>> >> To post to this list ....... discuss at ixda.org
>> >> List Guidelines ............ http://beta.ixda.org/guidelines
>> >> List Help .................. http://beta.ixda.org/help
>> >> Unsubscribe ................ http://beta.ixda.org/unsubscribe
>> >> Questions .................. list at ixda.org
>> >> Home ....................... http://beta.ixda.org
>> >>
>> >________________________________________________________________
>> >Welcome to the Interaction Design Association (IxDA)!
>> >To post to this list ....... discuss at ixda.org
>> >List Guidelines ............ http://beta.ixda.org/guidelines
>> >List Help .................. http://beta.ixda.org/help
>> >Unsubscribe ................ http://beta.ixda.org/unsubscribe
>> >Questions .................. list at ixda.org
>> >Home ....................... http://beta.ixda.org
>> >
>> >
>>
>

Kevin,

I have only been in the portal business for a year, but from day one
I approached it with a specific perspective. Users (I prefer
consumers) do not come to a site to be sold. The typically come for
some level of information or utility. That being said, if we build a
quality page with good value, then some level of advertising is very
likely acceptable. and, if you build it right, the people will come.
If the people are there, there is not a problem selling the adds.

The next step is to fix the nature of the adds. Often is not the ads
presence that is offensive, but the nature of the ads. Nobody want to
be insulted or annoyed. The dancing cowboys and in fact the majority
of run of schedule adds are not optimal. Adds that are contextual in
nature simply work better because of their relevance and will garner
more revenue per exposure.

Media agencies must get smarter about this... but I digress. I guess
I think in many situation the best we can do (designers, marketers,
and business working together) is to optimize the value of the ads
while minimizing the annoyance of them.

Mark

On Jul 18, 2007, at 6:14 PM, Kevin Doyle wrote:

> Hi Mark,
>
> Completely understood. I guess that it's not realistic for me to
> expect the Marketing- and Business-types to get the fact that users
> hate ads and will leave a page if they are flooded with new
> gimmicks to buy buy buy... just as it's probably not realistic for
> the Marketing- and Biz-types to get a user experience designer.
> It's so frustrating to have a well user-tested site all ready to go
> into final development... only to have a Marketing or Biz person to
> reject it because there's not enough on the page for advertisers.
> Heh, and I think that designers need something to whine and mope
> about... we're artists in the end, I think. Wait... maybe that's
> just me.
>
> Anyway, I think that you're on to a solution and you're totally
> right -- there is a sweet spot of Marketing/user-needs crossover,
> but it just takes time, money and someone with both of those to
> have to courage to fund the risk involved in researching and
> developing that solution...
>
> Kevin
>
>
> On 7/18/07, Mark Schraad <mschraad at mac.com> wrote:
> Hi Kevin,
>
> Totally get that, and I play the consumer/user advocate role
> arguing against ads nearly every day. But there is nearly always a
> business influence and the monetization is likely equally important
> if not more (to those making the final decision). My real point
> being that rather than have this decision forced on us, maybe we
> need to stop moping around and wining about it and accept it as a
> design constraint and create a solution that anticipates it.
>
> Mark
>
>
> On Wednesday, July 18, 2007, at 03:22PM, "Kevin Doyle"
> <kbdoyle at gmail.com> wrote:
> >I think that a pleasant user experience is good Marketing and
> supersedes any
> >"monetization" constraints. How many pages will a user click
> through if they
> >have to worry about a whole bunch of images, ads and Flash movies
> every time
> >they click "next". And pagination doesn't print well at all -- you
> have to
> >go to each page and print out things separately. You might be able
> to have a
> >"print-friendly version" of the article available, but I bet that
> once the
> >user clicks on the ad-free print-friendly, full article version,
> they're
> >going to finish reading the article there. Besides, with some good
> CSS
> >development, you don't even need to have a print-friendly verison
> of the
> >article available -- you just create a style sheet just for printing.
> >
> >On 7/18/07, Mark Schraad <mschraad at mac.com> wrote:
> >>
> >> I don't buy into the conspiracy thing here, if only in name.
> >>
> >> Monetization is a realistic and ethical design constraint.
> Designers need
> >> to understand how things (salaries and other resources) get paid
> for.
> >>
> >> For long lists, pagination makes perfect sense partly because it
> is so
> >> easy - and prints fairly well. User options for number of items
> to display
> >> are great.
> >>
> >> Mark
> >>
> >>
> >> On Wednesday, July 18, 2007, at 08:12AM, "Kevin Doyle" <
> kbdoyle at gmail.com>
> >> wrote:
> >> >Another Con:
> >> >It's not possible to print out an article that's spread out
> over several
> >> >pages unless there's an option to do so.
> >> >
> >> >I'm with Rob Brown on this one: pagination is a Marketing tool.
> I've
> >> found
> >> >pagination only to be useful on massive information returns
> like record
> >> and
> >> >internet searches.
> >> >
> >> >
> >> >On 7/18/07, Maxim Soloviev <maxim at deast.info> wrote:
> >> >>
> >> >> What about splitting long articles into few pages?
> >> >>
> >> >> Pro's:
> >> >> User always can create a bookmark on the page that he had to
> stop to
> >> read
> >> >>
> >> >> Con's:
> >> >> Someone who is going to read all article had to use pagination.
> >> >>
> >> >> --
> >> >> Maxim
> >> >> http://maxim.deast.info/
> >> >> ________________________________________________________________
> >> >> Welcome to the Interaction Design Association (IxDA)!
> >> >> To post to this list ....... discuss at ixda.org
> >> >> List Guidelines ............ http://beta.ixda.org/guidelines
> >> >> List Help .................. http://beta.ixda.org/help
> >> >> Unsubscribe ................ http://beta.ixda.org/unsubscribe
> >> >> Questions .................. list at ixda.org
> >> >> Home ....................... http://beta.ixda.org
> >> >>
> >> >________________________________________________________________
> >> >Welcome to the Interaction Design Association (IxDA)!
> >> >To post to this list ....... discuss at ixda.org
> >> >List Guidelines ............ http://beta.ixda.org/guidelines
> >> >List Help .................. http://beta.ixda.org/help
> >> >Unsubscribe ................ http://beta.ixda.org/unsubscribe
> >> >Questions .................. list at ixda.org
> >> >Home ....................... http://beta.ixda.org
> >> >
> >> >
> >>
> >
>

Sorry if I sound like a jerk here, but unless everyone here trusts Mr
Hoekman (which I do NOT), I wouldn't open that attached PDF file.

On 7/18/07, Robert Hoekman, Jr. <robert at rhjr.net> wrote:
>
> > My real point being that rather than have this decision forced on us,
> maybe we need to stop moping around and wining about it and accept it as a
> design constraint and create a solution that anticipates it.
> >
>
> Indeed. I think most of us would agree it's our job to consider
> business constraints and business needs when putting together a
> design. Lots of business make tons of money on ads, and there's no way
> around that. What we *can* do, though, is find better ways to
> integrate the ads, so they're less annoying and less intrusive.
>
> For example, I recently came up with the attached (click-through PDF),
> which I call "Ad Expand". I was sure Marketing would want more space,
> but they didn't. They loved the design. They thought it made the ad
> more noticeable and they understood that text ads work better than
> banners, so they liked the multi-step progression.
>
> I'm sure there are a billion other solutions. The point, obviously, is
> that it would behoove us to stop complaining and start devising better
> solutions. It's entirely possible users hate ads because of how
> they're typically implemented. Maybe there are ways to do it that
> don't suck. Maybe there are ways to advertise in a way that makes
> users happy.
>
> -r-
>
>

> Attached files are against the policies of this list in general:

Sorry - didn't realize that.

-r-

On Jul 17, 2007, at 3:32 PM, Jeffrey D. Gimzek wrote:

>
> i like best UI's that allow me to choose the pagination:
>
> ie: choose 20, 50, 100 items per page
>
> i am a power user with a fast connection, so i want 500 per page
>
> if you are normal shmuk with a dial up, you will want 20 per page

Not always. People with slow connections who are cognizant of
overhead will want more results per page. They know they may have to
wait twice as long to get ten times as many results, which may make
it worthwhile if they have to dig to find the results. More quality
time spent evaluating results, less spent waiting to get them in the
long run.

-- Jim