Infinite Scroll with "Load More" linking to Pagination?

[ichthyous]

Hello Webmasterworlders,

I am putting up a new WordPress site and the developer has implemented an infinite scroll on some of the category pages to load the custom posts. The problem is that I switched developers recently and the second one doesn't agree with the first one on best practice. The first one designed an infinite scroll to load ALL the posts on the first page of the category, which means a lot more links on one page but no need for Google to jump to page 2, 3, etc to index the rest of the content.

The second dev took over and "fixed" the first one's work while I was away...he made the load more button link to a new page entirely. So in effect the second method merges traditional pagination with lazy loading. The user can keep scrolling all the content on one page without clicking on pagination links, but Google will read until the last post, then it hits the load more button linking to page 2...for example:

mysite.com/photos/category/subcategory/
mysite.com/photos/category/subcategory/page/2/ - Only accessible from one link on the load more button at bottom of page 1
mysite.com/photos/category/subcategory/page/3/ - Only accessible from one link on the load more button at bottom of page 2
mysite.com/photos/category/subcategory/page/4/ - Only accessible from one link on the load more button at bottom of page 3

There are about 75 links per page including the header, but about 25 are nofollowed. So I would like your opinions about whether the first type or the second type of infinite scroll is better for getting content indexed and ranking in Google. Thanks!

[dipper]

option #3 - no infinite scroll at all.

But if you must, then refer to this doco .. [googlewebmastercentral.blogspot.com.au...]

[ichthyous]

I could turn regular old pagination on, but the first developer convinced me that it was better to use infinite scroll in terms of A) allowing people to scroll through more content more quickly and B) not diluting page rank by linking out to many other pages in the same category. Everything gets focused on one page. Considering that I may have 250 posts max for each category they won't be scrolling long to hit bottom. Now however this new dev is linking out to pages again. What would you recommend?

[dipper]

old style imho

[diddlydazz]

my two cents (pence)

any content that is hidden on page load *may* be devalued by the "powers that be"

in fact, there was a thread on WebmasterWorld (no time to locate at moment) which discussed whether the "hidden" content was actually indexed (for want of a better term, i mean searchable) by G

IMHO, pagination (done correctly) is the way to go

not diluting page rank

as i mentioned "pagination done correctly" - canonical tags

[aristotle]

Personally as a reader I much prefer everything on one page. It's annoying to have to keep clicking to the next page and waiting for everything to load. Sometimes I abandon the site and look elsewhere.

[diddlydazz]

aristotle

agreed from a user's perspective, if we really could concentrate on sites for "users" without triggering any adverse effects then life would be so much simpler!

the OP was asking "whether the first type or the second type of infinite scroll is better for getting content indexed and ranking in Google."

[diddlydazz]

here is the thread:

[webmasterworld.com...]

it was a while ago, seems like only yesterday!

[aristotle]

diddlydazz -- You're right about the OP, but I was responding to some posters in this thread, including you, who recommended pagination. In my view, not only is pagination annoying, it can also be a sign of fluff content.

[aristotle]

P.S. My advice to the OP is to just show everything on the initial pageload.

[ichthyous]

Thanks for the comments. So it's clear that some of you think that old school pagination works best, but frankly it doesn;t work all the well on my current site with 20+ pages of content. Google neither really indexes them and I'm sure that customers don' scroll through it all. The infinite load is much faster to browse a bigger selection of posts.

So, back to my original question....is the way the first dev set up the scrolling with all links loaded on one page better, even if it's 300 links --OR-- the second dev's solution which loads about 75 links at a time and a load more button at bottom that links to "Page 2,3,4,5" etc? The second solution really seems to offer pagination for Google while offering the convenience of loading content on one page for visitors.

Some people have mentioned using canonical tags with the second pagination setup. Does that mean that Google will index the pages and the content but only show PAGE 1 of the category in the serps? If anyone wants to see the dev site in action please send me a PM.

[diddlydazz]

You're right about the OP, but I was responding to some posters in this thread, including you, who recommended pagination

i agree pagination is annoying, absolutely, but as mentioned in the thread i linked to (and still true according to my tests), google does seem to view "hidden on load" content differently.

if the OP is referring to simply whether links in the "hidden" divs will get followed, then that is a different thing (as from my tests they do) - although it is hard to know if their value is affected by being in the hidden div

[ichthyous]

There are no hidden links on the page as is now. But how would the canonical tags come into play here? If we do end up staying with infinite scroll plus pagination should all the pages have cononical tags?

[ichthyous]

So personal observations about whether you like infinite scroll or not aside, any real insight into this issue?

[EditorialGuy]

Personally as a reader I much prefer everything on one page. It's annoying to have to keep clicking to the next page and waiting for everything to load.

A number of years ago, WIRED reported on a usability test that compared two versions of an article: a shorter version served up on one page, and a longer version that was divided among multiple pages. The results were intriguing: Most users thought the paginated article was shorter, even though it actually was longer. The majority of users also preferred the paginated version.

Pagination has several practical advantages:

1) Easier navigation within long text content such as articles.

2) If pages are created and organized by subtopic (rather than being chopped into chunks of a set length to fit a layout), each page is more focused than a huge catchall page would be. That's good for readers, and it should have SEO benefits as well.

Google has some useful advice on how to handle pagination at:

[googlewebmastercentral.blogspot.com...]

Finally: If pagination has earned a bad name in some circles, it's because too many Web sites crank out text-based slide shows or chop articles into bite-size chunks to get extra page views and ad impressions. This doesn't mean pagination is bad, it just means some Web sites are bad.

[ichthyous]

Thanks for the replies, but the original question was not pagination vs, infinite scroll...that issue has already been decided. The issue is WHICH FORM OF INFINITE SCROLL? If you read the original post I elaborate on the two methods two different devs have proposed and I am asking for opinions about that, not about pagination vs IS. Thanks!

[aristotle]

My recommendation is to simply show everything on the initial pageload. That means no additional clicking is required and nothing is initially hidden from view. This is the simplest way to do it codewise, and is also what most users would prefer.

[Robert Charlton]

I'll skip my negative comments about usability issues with the infinite scroll, except to say that "infinite" can turn into a very large number, which I gather the second method is intended to limit. I don't understand how that method incorporates traditional pagination, though (in ways that Google suggests, with rel="next" and rel="prev", etc).

Google's infinite scroll recommendations, which dipper links to, ought to be considered. (That said, I should add that I've not tried these)...

Infinite scroll search-friendly recommendations
Thursday, February 13, 2014
[googlewebmastercentral.blogspot.com...]


Another issue... one comment that caught my eye in your original post was this...

There are about 75 links per page including the header, but about 25 are nofollowed.

What is the nofollow intended to accomplish? 25 links out of 75 per page is essentially sending 1/3 of the internal PageRank of your site into a "black hole", as it's commonly described.

I would never use nofollow on internal navigation links. If they're outbound links, I wouldn't recommend having that many outbound links that I felt needed to be nofollowed (blog comments excepted, but that's a different situation, and I believe Google treats it differently).

[ichthyous]

I checked this link to Google's recommendations a second time and it appears that the second dev structured it in a similar fashion to Google's schematic. Ten images load at a time before the next set is auto loaded (no need to click a load more button), but no pagination links appear on the page. However when I check the page with a tool that spiders the page as Google it sees something different. The content is broken up into separate pages for each ten images, which are discoverable only if you type the url in directly. So it appears that the Show More link that is loaded at bottom is only visible to search engines, not the public since the public only sees what is auto loaded as they scroll. I am certain that this auto scroll method will get a lot more of my content seen by the user, but perhaps this odd way of hiding some links may get the pages penalized? Perhaps you want to take a look directly and let me know?

With regards to the no follow urls...perhaps my knowledge is outdated. I nofollowed links to pages that are not important to bringing in traffic like the contact us page and the search page etc. I want to concentrate all the inbound link juice from the home page to as few pages as possible, not spread it to pages that are not important for search. If this is an old way of thinking then it would be great to know and I will remove the nofollows and just let all pages get link juice. Thanks!

[aristotle]

I want to concentrate all the inbound link juice from the home page to as few pages as possible, not spread it to pages that are not important for search.

That's an old technique called "pagerank sculpting", which worked at one time, but no longer does due to a change that Googke made in how it calculates pagerank transfer. In fact, using nofollow tags on internal inks actually causes you to lose pagerank "juice", because in Google's current method a nofollow link acts as a sinkhole (or black hole) which absorbs pagerank juice and doesn't redistribute it.

So you need to make all internal links into dofollow links in order to conserve as much link juice as you can. Link juice will still "leak out" through external links, but there are things you can do to minimize that as well.

[ichthyous]

How quickly these techniques become outdated... Which is exactly what Google wants I think. Good thing I mentioned it here... I'll remove all the no follow tags asap

[aristotle]

ichthyous --
If you are loading a lot of images, it might be worthwhile to spend some time trying to reduce their file sizes. A few years ago I undertook a project to do this on all of my sites, and even though I had already done the same project once before, and thought the file sizes were already faily small, I was really surprized at how much additional size reduction I could achieve without producing a noticeable effect on what they look like on the screen.

[ichthyous]

Yes it's an image based site with large images at 1000px width / height. The larger sizes definitely help with sales, but slow the load time down. I am using photoshop to optimize and then WP Smush. I don't worry about users on slow connections as they aren't a target market for me.

[raseone]

After a lot of experimentation I actually reverted several things that were set-up with infinite ajax scrolling to regular, old pagination.

I found that the infinite scrolling would:

1.) allow pages to become HUGE in some cases.

2.) glitch out & ruin the whole experience if a single object had trouble loading or rendering or took too long.

3.) devalue content towards the bottom of the VERY long page.

4.) Provides the user with no impression of how long the page really is or if it ever ends.

Currently I have a mixed experience set up on one of my beta sites that provides an infinitely scrollable page that the user can jump through but this will surely give me a "duplicate content" and "cannonical links" issue with Google so I will probably just keep it old school & stick with what works.

[tangor]

Long scroll War and Peace and see how that works. (winkers) Actually works a treat as it is ONE item if it includes all.

Blog posts in infinite scroll, even by category, can be tedious and NOT SPECIFIC for search results. (scroll down 13 or 57 to get what the SE found).

Infinite has a use on text sites/comments. Most times those don't run to hundreds of thousands of words at a time.

Image sites, on the other hand, can grow painfully difficult, even on fast loading broadband. In that regard show thumbnails first, allowing a click through to the full size and rely on the user to back into the site to continue. Even then it just makes sense to limit to 25, 50, 75, 100 images (user selected).

Personally I am not a fan of infinite scroll, but I AM a fan of giving the user ALL of a text (think books, white papers, and reports, for example) all in one. The se's like that, the users certainly do and, though I know some will disagree with this statement, LIMITS my ad exposure --- but that's a different story. (and concept of doing business)

[raseone]

One little correction to my post...

In my mixed scroll/paginated version the urls are consistent either way but i still need to work on the possible perceived duplication issue from the initial landing url and the various urls once parameters are applied.

Its not actually hard to solve, just unfinished. I had to go back and look.

[aristotle]

tangor wrote:
Image sites, on the other hand, can grow painfully difficult, even on fast loading broadband. In that regard show thumbnails first, allowing a click through to the full size and rely on the user to back into the site to continue.

I agree with this approach, except instead of thumbnails, you should be able to use somewhat larger images. I've found that you can compress 300 pixel-width or so images to about 10 kb file sizes and still get reasonably good image quality on the screen. Then the user can click an image to see a large high-quality version in a popup window.

With this method you should be able to show everything on the initial pageload, without the complications of infinite scroll, as I recommended earlier in this thread.

[Robert Charlton]

Regarding nofollow and PageRank sculpting

How quickly these techniques become outdated...

Here's our announcement of Google's change in treatment of rel="nofollow", from six-and-a-half years ago. It's also first of many discussions we've had on the topic of (not) using nofollow internally...

Google Changes Treatment of PR 'Saved' by rel=nofollow Sculpting
June 2009
https://www.webmasterworld.com/google/3925952.htm [webmasterworld.com]

[ichthyous]

Thanks for the reminder Robert...keep in mind that not all of us live or die by Google algo changes so we aren't necessarily "up to speed". That's why we come here, for feedback from those who are!

[raseone]

I did the due diligence with rel=nofollow across several sites, on all outbound and low value internal links... gave it years to have an effect and there was no perceivable change in rank, traffic or indexing from google on any site.

Given the way google treats me im pretty sure taking my advice would be the worst possible thing you could do though.