Search Results Pagination (top & bottom?)

31 Dec 2009 - 12:43pm
4 years ago
9 replies
1971 reads
mcaskey
2008

Can I get some thoughts on pagination at the top of search results?

I haven't done any research around it, but my opinion this morning is
that pagination should only be at the bottom of the results set,
especially when the results set is taller than the viewport.

The context of pagination, in my mind, is only relevant when you've
reached the end of the results currently shown, and you want to see
more. Also, I think they clutter the space between the search box and
the results.

Am I missing something?

Mike Caskey

Comments

31 Dec 2009 - 1:43pm
Paul Sherman
2006

Don't forget about the case in which people are scanning a list or grid, but after scanning down to the bottom, they decide to go back and review items higher up in the list.

In this case - which I have seen play out many times in utesting of web-based apps - doubling up the pagination and sorting controls at top and bottom is the right design decision.

- - - - - - -
Paul Sherman, Principal, ShermanUX
User Experience Research | Design | Strategy
paul at ShermanUX.com
www.ShermanUX.com
+1.512.917.1942
- - - - - - -

On Dec 31, 2009, at 11:43 AM, Michael Caskey wrote:

Can I get some thoughts on pagination at the top of search results?

I haven't done any research around it, but my opinion this morning is that pagination should only be at the bottom of the results set, especially when the results set is taller than the viewport.

The context of pagination, in my mind, is only relevant when you've reached the end of the results currently shown, and you want to see more. Also, I think they clutter the space between the search box and the results.

Am I missing something?

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

31 Dec 2009 - 1:59pm
mcaskey
2008

I see. So the user who scrolls down, and then scrolls back up, would not have to scroll back down again if he decided to see the next page of results, provided the pagination were duplicated at the top.

I think in order to make this design decision I would need to measure the clutter/scroll tradeoff. :) How would you measure it?

Thanks!

Mike Caskey

On Dec 31, 2009, at 11:43 AM, Paul Sherman wrote:

> Don't forget about the case in which people are scanning a list or grid, but after scanning down to the bottom, they decide to go back and review items higher up in the list.
>
> In this case - which I have seen play out many times in utesting of web-based apps - doubling up the pagination and sorting controls at top and bottom is the right design decision.
>
>
> - - - - - - -
> Paul Sherman, Principal, ShermanUX
> User Experience Research | Design | Strategy
> paul at ShermanUX.com
> www.ShermanUX.com
> +1.512.917.1942
> - - - - - - -
>
> On Dec 31, 2009, at 11:43 AM, Michael Caskey wrote:
>
> Can I get some thoughts on pagination at the top of search results?
>
> I haven't done any research around it, but my opinion this morning is that pagination should only be at the bottom of the results set, especially when the results set is taller than the viewport.
>
> The context of pagination, in my mind, is only relevant when you've reached the end of the results currently shown, and you want to see more. Also, I think they clutter the space between the search box and the results.
>
> Am I missing something?
>
> Mike Caskey
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

31 Dec 2009 - 2:21pm
Paul Sherman
2006

Off the cuff I can think of a couple ways to operationalize a quick test.

But honestly, I probably wouldn't test it. There are plenty of good pagination design patterns in the wild that wouldn't add any significant clutter.

-Paul

- - - - - - -
Paul Sherman, Principal, ShermanUX
User Experience Research | Design | Strategy
paul at ShermanUX.com
www.ShermanUX.com
+1.512.917.1942
- - - - - - -

On Dec 31, 2009, at 12:59 PM, Michael Caskey wrote:

I see. So the user who scrolls down, and then scrolls back up, would not have to scroll back down again if he decided to see the next page of results, provided the pagination were duplicated at the top.

I think in order to make this design decision I would need to measure the clutter/scroll tradeoff. :) How would you measure it?

Thanks!

Mike Caskey

On Dec 31, 2009, at 11:43 AM, Paul Sherman wrote:

> Don't forget about the case in which people are scanning a list or grid, but after scanning down to the bottom, they decide to go back and review items higher up in the list.
>
> In this case - which I have seen play out many times in utesting of web-based apps - doubling up the pagination and sorting controls at top and bottom is the right design decision.
>
>
> - - - - - - -
> Paul Sherman, Principal, ShermanUX
> User Experience Research | Design | Strategy
> paul at ShermanUX.com
> www.ShermanUX.com
> +1.512.917.1942
> - - - - - - -
>
> On Dec 31, 2009, at 11:43 AM, Michael Caskey wrote:
>
> Can I get some thoughts on pagination at the top of search results?
>
> I haven't done any research around it, but my opinion this morning is that pagination should only be at the bottom of the results set, especially when the results set is taller than the viewport.
>
> The context of pagination, in my mind, is only relevant when you've reached the end of the results currently shown, and you want to see more. Also, I think they clutter the space between the search box and the results.
>
> Am I missing something?
>
> Mike Caskey
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

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

4 Jan 2010 - 10:30am
Erin Walsh
2007

Also keep in mind your sort options. If a very large result set is
returned and users are unable to adequately refine or sort, I've
often seem them jump to a middle page to see if that is closer to
their desired pinpoint. We often see this on real estate sites where
customers have entered a price of $0 or requested "Call for Price".
Rather than sift through pages of "Call for Price", I see users sort
by price and then randomly pick an interior result page to see the
price ranges on that page. This is all done before they scroll down
the page.

Wow, that sounded convoluted, my apologies if I confused anyone. You
may have complete control over the refine and sort, but we often hit
business rules that hinder the optimal search and result features.

Thanks,
Erin

erin walsh | product developer | For Rent Media Solutions™
150 granby street, 16th floor | norfolk, va 23510
p:757.351.8444 | f:757.961.4827
erin.walsh at ForRent.com| www.FRMediaSolutions.com

You Have Multiple Marketing Needs... We Have Multiple Solutions!

On Dec 31, 2009, at 2:21 PM, Paul Sherman wrote:

Off the cuff I can think of a couple ways to operationalize a quick
test.

But honestly, I probably wouldn't test it. There are plenty of good
pagination design patterns in the wild that wouldn't add any
significant clutter.

-Paul

- - - - - - -
Paul Sherman, Principal, ShermanUX
User Experience Research | Design | Strategy
paul at ShermanUX.com
www.ShermanUX.com
+1.512.917.1942
- - - - - - -

On Dec 31, 2009, at 12:59 PM, Michael Caskey wrote:

I see. So the user who scrolls down, and then scrolls back up, would
not have to scroll back down again if he decided to see the next page
of results, provided the pagination were duplicated at the top.

I think in order to make this design decision I would need to measure
the clutter/scroll tradeoff. :) How would you measure it?

Thanks!

Mike Caskey

On Dec 31, 2009, at 11:43 AM, Paul Sherman wrote:

> Don't forget about the case in which people are scanning a list or
> grid, but after scanning down to the bottom, they decide to go back
> and review items higher up in the list.
>
> In this case - which I have seen play out many times in utesting of
> web-based apps - doubling up the pagination and sorting controls at
> top and bottom is the right design decision.
>
>
> - - - - - - -
> Paul Sherman, Principal, ShermanUX
> User Experience Research | Design | Strategy
> paul at ShermanUX.com
> www.ShermanUX.com
> +1.512.917.1942
> - - - - - - -
>
> On Dec 31, 2009, at 11:43 AM, Michael Caskey wrote:
>
> Can I get some thoughts on pagination at the top of search results?
>
> I haven't done any research around it, but my opinion this morning
> is that pagination should only be at the bottom of the results set,
> especially when the results set is taller than the viewport.
>
> The context of pagination, in my mind, is only relevant when you've
> reached the end of the results currently shown, and you want to see
> more. Also, I think they clutter the space between the search box
> and the results.
>
> Am I missing something?
>
> Mike Caskey
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

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

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

4 Jan 2010 - 1:23pm
msweeny
2006

Good Morning All,

I'm fascinated by the discussion around search results pagination that is
used by less than 1% if the users (unless you are talking about enterprise
search in which case you might dip a toe on the sunnier side of 1%. WE all
know that searchers rarely go past page 1 let alone past the top 5 results.
So, the design for page 1 would be a more critical issue than where to put
access to page 2 and so on.

Just a thought...

marianne
Daedalus Information Systems

-----Original Message-----
From: discuss-bounces at lists.interactiondesigners.com
[mailto:discuss-bounces at lists.interactiondesigners.com] On Behalf Of Erin
Walsh
Sent: Monday, January 04, 2010 7:31 AM
To: IXDA list
Subject: Re: [IxDA Discuss] Search Results Pagination (top & bottom?)

Also keep in mind your sort options. If a very large result set is
returned and users are unable to adequately refine or sort, I've
often seem them jump to a middle page to see if that is closer to
their desired pinpoint. We often see this on real estate sites where
customers have entered a price of $0 or requested "Call for Price".
Rather than sift through pages of "Call for Price", I see users sort
by price and then randomly pick an interior result page to see the
price ranges on that page. This is all done before they scroll down
the page.

Wow, that sounded convoluted, my apologies if I confused anyone. You
may have complete control over the refine and sort, but we often hit
business rules that hinder the optimal search and result features.

Thanks,
Erin

erin walsh | product developer | For Rent Media SolutionsT
150 granby street, 16th floor | norfolk, va 23510
p:757.351.8444 | f:757.961.4827
erin.walsh at ForRent.com| www.FRMediaSolutions.com

You Have Multiple Marketing Needs... We Have Multiple Solutions!

On Dec 31, 2009, at 2:21 PM, Paul Sherman wrote:

Off the cuff I can think of a couple ways to operationalize a quick
test.

But honestly, I probably wouldn't test it. There are plenty of good
pagination design patterns in the wild that wouldn't add any
significant clutter.

-Paul

- - - - - - -
Paul Sherman, Principal, ShermanUX
User Experience Research | Design | Strategy
paul at ShermanUX.com
www.ShermanUX.com
+1.512.917.1942
- - - - - - -

On Dec 31, 2009, at 12:59 PM, Michael Caskey wrote:

I see. So the user who scrolls down, and then scrolls back up, would
not have to scroll back down again if he decided to see the next page
of results, provided the pagination were duplicated at the top.

I think in order to make this design decision I would need to measure
the clutter/scroll tradeoff. :) How would you measure it?

Thanks!

Mike Caskey

On Dec 31, 2009, at 11:43 AM, Paul Sherman wrote:

> Don't forget about the case in which people are scanning a list or
> grid, but after scanning down to the bottom, they decide to go back
> and review items higher up in the list.
>
> In this case - which I have seen play out many times in utesting of
> web-based apps - doubling up the pagination and sorting controls at
> top and bottom is the right design decision.
>
>
> - - - - - - -
> Paul Sherman, Principal, ShermanUX
> User Experience Research | Design | Strategy
> paul at ShermanUX.com
> www.ShermanUX.com
> +1.512.917.1942
> - - - - - - -
>
> On Dec 31, 2009, at 11:43 AM, Michael Caskey wrote:
>
> Can I get some thoughts on pagination at the top of search results?
>
> I haven't done any research around it, but my opinion this morning
> is that pagination should only be at the bottom of the results set,
> especially when the results set is taller than the viewport.
>
> The context of pagination, in my mind, is only relevant when you've
> reached the end of the results currently shown, and you want to see
> more. Also, I think they clutter the space between the search box
> and the results.
>
> Am I missing something?
>
> Mike Caskey
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

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

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

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

4 Jan 2010 - 2:38pm
mcaskey
2008

Hey Marianne,

I agree that the content of the first page of results is most important, since the first page of search results is the only one that is likely to be used.

Do we agree that pagination controls usually should only appear at the end of the first page of results, rather than cluttering the page with controls that aren't likely to be used?

Thanks!

Mike C.

On Jan 4, 2010, at 11:23 AM, marianne wrote:

> Good Morning All,
>
> I'm fascinated by the discussion around search results pagination that is
> used by less than 1% if the users (unless you are talking about enterprise
> search in which case you might dip a toe on the sunnier side of 1%. WE all
> know that searchers rarely go past page 1 let alone past the top 5 results.
> So, the design for page 1 would be a more critical issue than where to put
> access to page 2 and so on.
>
> Just a thought...
>
> marianne
> Daedalus Information Systems
>
> -----Original Message-----
> From: discuss-bounces at lists.interactiondesigners.com
> [mailto:discuss-bounces at lists.interactiondesigners.com] On Behalf Of Erin
> Walsh
> Sent: Monday, January 04, 2010 7:31 AM
> To: IXDA list
> Subject: Re: [IxDA Discuss] Search Results Pagination (top & bottom?)
>
> Also keep in mind your sort options. If a very large result set is
> returned and users are unable to adequately refine or sort, I've
> often seem them jump to a middle page to see if that is closer to
> their desired pinpoint. We often see this on real estate sites where
> customers have entered a price of $0 or requested "Call for Price".
> Rather than sift through pages of "Call for Price", I see users sort
> by price and then randomly pick an interior result page to see the
> price ranges on that page. This is all done before they scroll down
> the page.
>
> Wow, that sounded convoluted, my apologies if I confused anyone. You
> may have complete control over the refine and sort, but we often hit
> business rules that hinder the optimal search and result features.
>
> Thanks,
> Erin
>
> erin walsh | product developer | For Rent Media SolutionsT
> 150 granby street, 16th floor | norfolk, va 23510
> p:757.351.8444 | f:757.961.4827
> erin.walsh at ForRent.com| www.FRMediaSolutions.com
>
>
> You Have Multiple Marketing Needs... We Have Multiple Solutions!
>
>
>
> On Dec 31, 2009, at 2:21 PM, Paul Sherman wrote:
>
> Off the cuff I can think of a couple ways to operationalize a quick
> test.
>
> But honestly, I probably wouldn't test it. There are plenty of good
> pagination design patterns in the wild that wouldn't add any
> significant clutter.
>
> -Paul
>
>
>
> - - - - - - -
> Paul Sherman, Principal, ShermanUX
> User Experience Research | Design | Strategy
> paul at ShermanUX.com
> www.ShermanUX.com
> +1.512.917.1942
> - - - - - - -
>
> On Dec 31, 2009, at 12:59 PM, Michael Caskey wrote:
>
> I see. So the user who scrolls down, and then scrolls back up, would
> not have to scroll back down again if he decided to see the next page
> of results, provided the pagination were duplicated at the top.
>
> I think in order to make this design decision I would need to measure
> the clutter/scroll tradeoff. :) How would you measure it?
>
> Thanks!
>
> Mike Caskey
>
>
> On Dec 31, 2009, at 11:43 AM, Paul Sherman wrote:
>
>> Don't forget about the case in which people are scanning a list or
>> grid, but after scanning down to the bottom, they decide to go back
>> and review items higher up in the list.
>>
>> In this case - which I have seen play out many times in utesting of
>> web-based apps - doubling up the pagination and sorting controls at
>> top and bottom is the right design decision.
>>
>>
>> - - - - - - -
>> Paul Sherman, Principal, ShermanUX
>> User Experience Research | Design | Strategy
>> paul at ShermanUX.com
>> www.ShermanUX.com
>> +1.512.917.1942
>> - - - - - - -
>>
>> On Dec 31, 2009, at 11:43 AM, Michael Caskey wrote:
>>
>> Can I get some thoughts on pagination at the top of search results?
>>
>> I haven't done any research around it, but my opinion this morning
>> is that pagination should only be at the bottom of the results set,
>> especially when the results set is taller than the viewport.
>>
>> The context of pagination, in my mind, is only relevant when you've
>> reached the end of the results currently shown, and you want to see
>> more. Also, I think they clutter the space between the search box
>> and the results.
>>
>> Am I missing something?
>>
>> Mike Caskey
>> ________________________________________________________________
>> Welcome to the Interaction Design Association (IxDA)!
>> To post to this list ....... discuss at ixda.org
>> Unsubscribe ................ http://www.ixda.org/unsubscribe
>> List Guidelines ............ http://www.ixda.org/guidelines
>> List Help .................. http://www.ixda.org/help
>>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help

4 Jan 2010 - 3:03pm
Andy Edmonds
2004

I'd generally agree that real estate at the top of the page tends to be too
valuable to spend on pagination.

Some points:

- In e-commerce, pagination tends to be used to a greater degree than the
low/1% stat from web search
- The #1 use case I've seen for top of page pagination, while working at
MSN Search, was to return to page 1 from page 2 after noticing the results
were less relevant or uninteresting. The back button works for this too and
the controls were little used aside from this situation. We dropped them.

I've actually tested top of page pagination in an e-commerce venue by
creating a control condition in which the top of page copy was increased in
size to maintain the same vertical footprint in a multivariate design that
toggled top of page pagination. We found that top pagination added little
value to this situation, while screen real estate added consistent value.
The copy seemed to effect only a subset of users and thus while positive was
not strong enough to warrant giving up vertical real estate.

One deep thought on pagination to close: In a talk at Web 2.0 Expo in 2007 I
showed a graph from MSN Search in which users click on the numeric links in
standard < 1 2 3 4 5 > style pagination to a decreasing amount as the page
number increases. I attribute this to user goal differences, "I'm going to
check page 2" versus "I'm going to scan these results".

hth, Andy

On Mon, Jan 4, 2010 at 11:38 AM, Michael Caskey <lists at casadev.com> wrote:

> Hey Marianne,
>
> I agree that the content of the first page of results is most important,
> since the first page of search results is the only one that is likely to be
> used.
>
> Do we agree that pagination controls usually should only appear at the end
> of the first page of results, rather than cluttering the page with controls
> that aren't likely to be used?
>
> Thanks!
>
> Mike C.
> <snip>
>

4 Jan 2010 - 5:03pm
Nancy
2009

Our site has a lot of data and I have Search on the top of the page
and pagination only at the bottom. The reason for doing this was that
the search on the top searches for alphanumeric characters across all
the columns and shrinks the numbers of row to have corresponding
value. Now, if this the case users don't generally have to go to
different pages and therefore pagination on the top and bottom would
be too much.
Also there is number of results per page drop down option to reduce
scanning too many pages.
Hope this helps!

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=48156

6 Jan 2010 - 8:04pm
DampeS8N
2008

Pay EXTRA close attention to what you are paginating. Sometimes a
big-ol-list of items page after page is the wrong fit. In the app I
am working on, for example, there is a big old list of all the
articles released in the last forever. The obvious thing to do was to
paginate them into 10 or 20 item chunks, but what is the use of such a
list. really?

We decided that our users were using this mainly to get a handle on
what everyone was doing that day, and to scan back to items they knew
were released recently but weren't sure what they were named.

So we paginated by day. However many came out that day, are shown.
The idea being that should there ever be THAT many items on a single
day, we'll look into ways to filter it, but for now there are only
10-30 articles released a day, so it isn't a big deal.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=48156

Syndicate content Get the feed