Many of the large content and e-commerce sites we've worked with experience a disease I like to call "page bloat." Symptoms include pagination of content pages, creation of new pages that simply provide alternate navigation methods and site architecture design that follows the little-known usability rule from well-known guru, Wrongy McLovestoClick - "more pages are always better than fewer pages."
I'm firmly in the camp against Mr. McLovestoClick - my idea of the perfect user experience is to deliver the content a user wants in the fewest possible clicks from the page they arrive on. This means making navigation simple, direct and obvious. It doesn't mean creating thousands of paths for a user to follow. Another rarely invoked usability rule that I do agree with:
More Choices with Exactly What the User Wants < Fewer Choices with a Reasonably Decent Selection
What? The user would rather settle for something imperfect then have to wade through dozens of sidebar navigation units to find their exact match? Yep. Ever wonder why those top 3 results at the search engines received 70% of clicks, even when search results used to suck - it's because users are happier compromising now than spending time getting it perfect.
Most folks agree on these basic truths about web surfing, but I know some of you out there are still saying, "But Rand, my business model is based on page views - I need to maximize the number of clicks to get the most revenue from my traffic." I still say eliminating unneccessary pages, killing paginated articles and removing extraneous navigation will help. A better user experience means more repeat visits, more links, more viral sharing and ultimately, more page views. In the long run, who's going to win the traffic game? The site that delivers what users want, or the site that makes you click 12 times to get what you could have found elsewhere with 2 clicks?
That's not all the problems with page bloat - they're bad for search engine indexing and ranking, too. Just have a look:
I invite you to review your own sites and ask if you need to provide 3 or 4 navigation systems when 2 are used by 95% of visitors. Do some testing to see if those content pages earn more links and more traffic when they're paginated or when they provide all the content at once (not - pagination can be OK, but a lot of sites overdo it). Look at the click-paths of your visitors through analytics and ask yourself - could I have gotten them there faster?
And think of the poor spiders! They have so many billions of pages to crawl already - do you really need to add to that burden?
Hi everyone-
First, I want to say that I am a Rand fan. I think he should write a book. Plus I hate blogs. It's a big deal when you see me post to a blog. Please do not take the following comments personally.
Look, I understand that neither search engines nor end users want redundant content delivery. Not disputing that.
What I am disputing is this 3-click rule and the difference between actual and perceived clickstreams. Jared Spool (or was it Nielsen) disproved the number of clicks thing a long time ago. As long as the clicks provide a strong information scent (it was Spool), as long as visitors believe they are making progress, then people will click more than 3 times.
I've confirmed Jared's findings many times over with my own client usability test results and clickstream analyses.
I also dispute the pagination of content opinions. Keep in mind, these are your opinions and not the results of actual user tests. If you find people prefer scrolling over clicking, then okay. Go for it. And if you find people prefer clicking over scrolling, then go for that. But don't assume that your target audience shares your opinion. You have test-measure-test-measure. And I believe that most SEOS have little or no training/experience/formal education in user-centered design.
There are reasons to break up content into multiple pages (usability and information retrieval reasons), and reasons not to do so.
People do scroll, and people do click. I see it all of the time. I test it all of the time. So do the search engines.
I don't mean to sound rude or condescending. What I'm reading? Seems to be opinions rather than analyses of genuine search behavior on users who fit personas and profiles. I'd rather read information on that.
My 2 cents on the topic.
I've found it to be wildly dependent on the audience. My main client has very unsavvy users, and testing indicates that they tend to prefer hierarchical structures and even an extra click or two to being confused by a lot of content on a single page or open-ended navigation (such as a sitewide search). The "right" answer is often very context dependent. That said, I don't dispute Rand's assertion that page bloat is generally a bad thing. As sites grow, we tend to add content without thinking about how that new content affects the information architecture. The "right" IA might be different for different sites, but adding content willy-nilly without thinking it through is always a bad thing.
Shari - my suggestions aren't contradicting anything you're saying. I agree that there can be cases where pagination of content is OK (though I'd say it's very rare). I also agree with your test/measure philosophy - it's a cornerstone of all good SEO and conversion rate improvement.
I'm pretty big on the 3-clicks - both for engines and users, but again, not universally. I think making content accessible to spiders and to humans in a logical fashion that is "as fast as possible" and "as few clicks as possible" is the key. That doesn't always mean 3 or fewer, but my perspective is the smaller the better (unless, as I noted in the post, it's causing navigation and page bloat).
Thanks for your contributions - they've made this blog entry even more valuable.
p.s. Sorry to hear you don't like blogs, but thrilled to have you contributing here.
i completeley agree, bether have 5 pages with good information.than 15 with the same duplicated content..thanks for the article i love it!
The googlebot development continues... That thing looks pretty mean.
Agree with you completely Rand - I hate pagination of content. That's what scroll wheels are for ;)
The google spider of death, the stuff nightmares are made of.
Would the scroll wheel be an enemy of the spider?
Ahh! GoogleSpider (Crab?) looks like some horrific beastie from The Dark Crystal!
I needed to google the film, but image search came up good - you are absolutely right. Someone needs to buy one of these for a whiteboard Friday: Garthim.
When designing for our clients, we always try to adhere to the old 3 click rule for finding anything on a website. I absolutely hate clicking multiple times just to find what I am looking for, so why would I design a site any other way.
By the way, had a hard time connecting earlier, glad to see the site back.
I agree, Classa. I try to follow the same guideline with my development as well; I think it makes everyone (site's owners AND users) focus on what content is truly important.
3 good reasons for pagination of content pages
1. Page Usability. Some articles are big. As in reading time will take +10 minutes. In these cases, you will want to break up long articles into readable bite-sized chunks. No one wants to wait for a 20,000 word, image-heavy, table-heavy document to load.
2. Retrieval Time / Server Load. For the 5 bajillion properties retrieved if you searched the NYC metro for a new apartment, you'll want pagination, not just for users but also for less stress on your backend.
3. $$$$. Pagination = more pages viewed.
More pages viewed = more advertising revenue.
If you're in the top tier of eCPM earners, your media sales team is someone you need to listen to. If media sales say that you can make an extra $5 million from pagination AND your usability team says it can be done without screwing with user satisfaction, guess what? Pagination is a go.
What's a solution for spiders and consumers that hate multiple pages? Offer a 'single page' option. For example, pagination on a news article might look like:
[Page: 1, 2, 3, 4, Single Page]
So even if a user clicks on the single page format, 2 page views are better than 1 right? $_$ (dollar smiley)
There is also nothing wrong with alternate navigation schemes if it enhances your user clickstream. A good example would be any type of directory, archive or widget that shows "Top 100 most popular articles". This type of page is more than likely to drive extra reach without cannabilizing existing usage flow.
If you're afraid of duplicate content, you can noindex or disallow robot crawling to the duplicated sections.
Test, test, test!
Here are a couple more for you shor....
4. Greater Keyword Coverage - the additional pages can be SEOed for additional search terms.
5. $$$$. Pagination = more ads in premium position and higher yield per visitor.
6. More and Better Targeted Contextual Ads - The greater keyword coverage - on separate pages gives you tighter ad targeting and higher CTR - even on same page.
I have found that by using careful analytics, that some sites will make a lot more money from contextual ads when a 2000 word article is broken down into about five keyword focused pages.
Reasons against
I'm not always 100% against paginated content, but for the vast majority of sites that use it on content pages, it's completely overdone and hurts more than it helps (except for those folks who leverage it for page view impressions with advertisers).
I agree that for most sites pagination hurts user experience. but sometimes we've decided to sacrifice better usability and search friendlier pages for CPM revenue and less server load. Compromising seems to be a half the job for inhouse search marketers ;)
For users that are looking for pagination guidelines, check out the Yahoo developer library's pagination resource.
Insightful post as always. My company hosts a similar site compared to our competitor, but our content is structured so full articles can be viewed by scrolling down. The articles on our nemesis's site warrants continued clicking encompassing several pages in order to get through an entire article (which I think is a laborious endeavor). I imagine more people rather have the information all on one page with the possibility of linking to germane articles somewhere on the page.
Speaking of nemeses...Exclusively in Moz theatres this summer... Googlebot vs. Googlevenom: The Battle for the SEO Grail
"...this film is not currently rated..."
Speaking of nemeses...Exclusively in Moz theatres this summer... Googlebot vs. Googlevenom: The Battle for the SEO Grail
"...this film is not currently rated..."
GAMURA! SAVE US! (oooohhhh I wonder how many are this geeky with Godzilla movies *grins*)
orange juice burns out your nose :P, Thanks deCabbit
And makes a sticky mess.
*tips virtual hat*
My work here is done....
*rides virtually off into virtual sunset, virtual cheesy western movie music virtually playing in virtual background (virtually)*
frightening spide :d
indeed fewer clicks are better.
in order to increase navigation i put on one site the search button below the fold. once i've moved it back up, the sales incresed...
fewer clicks drive more happy customers!
i'm not sure that you can assign the increase in sales only to the move of the search button.
but! i agree that nobody wants to click just for the sake of it. especially if we're talking about an online store!
Absolutely agree with you on the pagination and too much navigation choices Rand.
When I read news sites or articles online that require me to click on "page 2" or "next" to continue reading annoys the hell outta me.
Having too many different navigation methods leads to confusion and the visitor being overwhelmed with where to start.
BTW, love the Google Spider!
Interesting Blog at a time when i am redesigning a site which has "hundreds and hundreds of pages" using the same "pagination" you talked about.
the way i have gone about is add an "archives" section where after each day the previous day's posts (content) goes in there and the user can select the date and/or month to see those results thus reducing those mindless pagination horrors.
so far in the inhouse testing which has been done, the amount of time an user spent to search for a particular info is cut by almost 50%
Good Post Rand :)
I totaly agree this is a great post I hate not finding what I want when I want it. I am digging this post asap.
Rand, your article is dead on. I've seen this quite a few times. In one case, we had a new client and his previous SEO's had added some truly useless pages to their site. I immediately spotted this and convinced the client to let us remove them. When we deleted those pages, their rankings in Google, across the board, literally shot up the next day. Many pages rose 10-20 spots, while one keyword went from 150 to 9. (Background: This site had been around for several years, had medium to decent rankings - not great, and not consistent - and wasn't sandboxed.)
This was actually a very timely blog.
I was able to shunt almost all the javascript to an external .js file and it did wonders. The CSS was already external also. I however, have some concerns.
The web page also has the same navigation bars on every page, but the graphics and javascript is contained in the cells of the same table that everything else is in. Is there really a way to replace totally replace a table with just CSS? And could the javascript be exported to the aforementioned file? Or just leave well enough alone?
Here is the website if anybody wants to see it.
-Clif
You could quite easily replace that table/js based nav bar with a css based one. Have a look at https://css.maxdesign.com.au/listamatic/ for some basic examples (and good place to start off from)
Thanks, deadmoon, I appreciate it.
I am also a big fan of the 3 click rule for SEO, a sitemap with text links to all category pages will fix most shopping sites A/ it makes it easier for the bots B/ you get another anchor text link for each of the category pages
Very interesting food for thought. I definitely noticed that on one of our sites. This is exactly what I have been trying to remedy since I started. Get the useful info to the user quick and they will come back more ofter. That's the gist I get from it. Good article!
let's face it...the attention span of web surfers are notoriously short...one of the reasons why we always make sure client sites upload as fast as possible. Then the content issue, and navigating to said content gets raised. One needs to draw a balance between having lots of unique, interesting content that is easy to navigate to, and the dreaded page bloat...
Good post. A lot of people get in the mindset of "the more content the better" and scatter useless content all over to try to increase rankings.
Its an old article but I feel strongly about this topic.
I am ok with creating more pages on the website as far as it adds value. However, for each customer - figure out what they are here for and how you get them to their desired solution as soon as possible.
As I'm finding with an e-commerce client, multiple paths also tends to create duplicate content, in the eyes of the spiders. We have 32,000 pages indexed, most of which are alternative search results of one form or another. It is a rich, database-driven site, but the primary content that we would want to link to (and that Google would also prefer to link to) is more like 2,000 pages.
BTW, if I ever get them out of Supplemental Hell, there's a big fat YouMoz post coming.
I spend more time "tuning" content that is already on the site than I spend creating new content. If a person is serious about maximizing the yield from each piece of content it is very important to test. The important metric will be "average income per visitor" or "number of sign-ups for 1000 visitors" or whatever your key performance indicator might be.
That should be your guide rather than a general rule.
In this debate about breaking an article up into multiple pages I think that there are many different ways to link those pages and the choice among those methods will be extremely important in how successful your visitors will be in moving from one page to another. Poor performance of multipage articles might be related back to the method by which the pages are linked. If analytics allows you to see that visitors are going page by page through the article or jumping to the most popular "chapters" we might find that either of those methods can be a success. (who says that visitors want to read in a linear manner through your article and if you gave them the chance to jump to the "meat" you might be surprised at how many of the drop offs did that instead of suffering or scrolling over a long long page.
My firm belief after reading this thread is that analytics is the only answer to assessing onsite performance as defined by your personal KPI - and maintaining that your KPI might not be 100% in agreement with why the visitor came to your site.
The take away for me is to hear Rand say that linkability went up by combining the pages. That is the ultimate KPI and I'll sacrifice current income to get it.
EGOL - Of course, the problem is determining the ROI and value of a link.. Very, very tough to do.
Most of the time, it doesn't take very many links to make a big difference... So, I might toss the long edition of a couple of old articles back up and see if they get linked.
Once again, an example of critical elements that often don't get thought out enough. To often the CMS or "this is how we always handle this" becomes the driving factor.
But great points, on both sides of the fence, for or against. The real deciding factor comes back to the visitors and what is best for them, whether they be living and breathing beings or that quite frightening Google spider thing!!
And these aren't always the same.
I think the biggest takeaway from this post and the comments is a reminder that the development and interface of a website should be a very conscious activitiy. So much of this becomes routinue, old hat that it is easy to forget the importance and forget what a vital role it plays in both usability, user satisfaction, and SEO.
A lot of news sites have horrible pagination, some have over 8 or 9 pages for an article. A pain to use, and if they don't offer just one page for printing, even more of a pain to print off.
The biggest thing i love from the new web2.0 cliche'd sites (not the myspace sorts of web2, the ones full of gradients etc) is that they're often (although, not always!) really easy to find what page you're looking for.
Better call Kristin Dunst, there is another spidey on the loose!
I still say eliminating unneccessary pages, killing paginated articles and removing extraneous navigation will help.
I absolutely agree that paginating articles is the wrong way to go. Rand you have long posts, but if they were split over 4 pages there is no way i would get through the article. We like to read on the web like newspapers and scan for content.
By extraneous navigation i assume you mean things like 'tags'. Navigation such as breadcrumb trails can be very helpful, especially if a user has clicked into a deeplink on your site.
haha Google looks mean, great blog item!
Google crawler is scary... I'm going to nightmares for weeks where I am running through my site trying to escape the spider.
*shudders*
Good to see these thoughts codified. If I can cite a source that isn't just my brain, it helps others believe me!
Now... the challenge of anchor text in ranking *girds self*
I totally agree with your points, its a nightmare having to click next page again and again.
I am in the middle of a redesign to my site webtrendz.co.uk and i have many pages here which will be culled and many that need to be improved.
thanks for the assisted insight.
SEOmoz uses pagnation at the bottom of the main blog pages ...
So, are you planning on removing the 114 pages that use numbers for anchor text and slimming it down or what? Aren't they there for crawlers and users to find deep level content?
I agree, don't get me wrong, but i think you should take it on a case by case scenario as to whether multiple navigation options are needed to find deep content pages or products.
On this blog, I can search with the search bar, view categories, related posts, popular posts, and comments. However, there is no archive of posts and no other ways for me to find content, except the page links at the bottom.
Have your findings convinced you to move those links or replace them with some other method like an archived search?
eCopt - if you look, those paginated results are just for people. Search engines don't get to spider them. It's good for those who might want to read tons and tons of blog entries or are looking to find one from a few weeks back but can't remember the name or category.
Also - no archive? The categories contain all the blog posts - they serve as a semantic archive.
Thanks for the clarification on being for users. I know I use them, so I was worried they were going bye bye!
"The categories contain all the blog posts - they serve as a semantic archive."
Well sure, there's also an author archive if you think of it in those terms. I guess I was thinking like more of traditional WP blog archive by year, month and day in asc or des order.
I think semantic archives by categories is much more usable. When was the last time you came upon a new site and thought "Let me see what were they had to say in November 2006"? More often it is "Let me see what they had to say about SEO", for example.
I agree with you. One of our platforms has some nice navigation features that may not be used by many, but that are probably causing some duplicate content issues...
when you gave the stanford publishing course and went over this issue, it opened a lot of eyes in my company -- so kudos to you!
Great post and I totally agree with you. User experience is the biggest aspect of a website that should be really thought out. All one has to do is remember that most of the web surfers today are very impaitent and want what they are looking for in the quickest time in easiest way possible. The longer or more difficult it is for them to find what they want on your site increases the chance that they will exit out and go to another site.
And your Google "spider" scares me.
I wouldn't mind seeing an example of what you had in mind Rand...maybe a link?
Couldn't agree more with you Rand!
I agree with this article 100% , I am glad to actually see that someone else has my views.
Just the other day I was talking with my wife about , that website about.com I mean everypage is like what you said , 50 words with 200 links to the rest of the article..
I make long pages, I rather have the visitor be happy nad come back, then make an extra 20 cents on a click
here is an example of a page of mine
https://www.plumberhelper.com/pool-and-spa/spa-buyers-guide/emerald-spas.html
If you look you'll see how I have a nice article with no bs, no continue links into 20 pages.
I think the visitor will get the most out of the site and maybe book,mark it and come back :)