Pagination sucks. Chronological, topical archives rule

Pagination sucks for SEO. Chronological archives work way better. In this article I'll tell you why it sucks and how to deal with it.

Note: If your website is small (10-1000 articles), then this story is not relevant for you.

But if you have a large website, with thousands, hundred thousands articles, then maybe you have noticed that your old articles don't rank really well. If you haven't: that is most probably the case. Dive into analytics and look it up ;)

Odds are that this is caused by a pagination system.

Why pagination is evil and should be banned

I'll try to keep it short:

  • the 'content' on pagination pages (containing usually not much more than duplications (worst), excerpts (bad) or at best only references)) continually moves to other URLs. A new article is written and enters page 1 of the pagination system at the top. The bottom article of page 1 moves to the top of page 2, et cetera. The consequence for Google is that a huge number of URLs continuously has both duplicated and moving content. Google attempted to solve this problem in 2011 by introducing the rel='next' and rel='prev' attributes. This works, for small sites.
  • do you know the saying: 'if you want to hide a dead body, send it to page 2 on Google'? This also applies to your own website. Older articles are harder to reach for Googlebot. They do not receive fresh link juice anymore. Only other old articles link to them.
  • pagination pages usually lack context. Articles are put together on the same page because they happen to be written on a certain date. That is fine if your website always writes about the same subject. But if your website covers multiple subjects, you should want to group topically related articles together.

This could be by design: if you're a newspaper, you usually concentrate on the news, and not on old subjects (which is a mistake, but that is food for another article). But if you want your old articles to remain fresh, topically interconnected, and get traffic, then you should combat the pagination problem.

So why should pagination be banned? Because it sucks. See above.

Why 'rel=next' and 'rel=prev' are not a solution

Most SEO-specialists tell you to start using the rel='prev' and rel='next' attributes. But ... that is a workaround and not a solution.

It is a workaround because your older articles are still buried in your pagination system. And they're not topically interconnected.

So, what should be done?

There is a better way, and it doesn't even require a lot of work. Because this is what works better:

1. Chronological archives

A 'chronological archive' is not a pagination page, but a list of articles that was written in a certain period, for instance in a year (like I do on this very blog). The number of references on the page will of course vary, but the fact is that when an article is written in a certain year, it will always remain available on that same archive URL. No fuss with continuously moving content anymore! That solves one part of the pagination problem.

P.s.: also change your archive template

Besides that, it is always a good idea to make sure that your archive pages (either chronological or paginated, doesn't even matter) are what they are meant to be: a placeholder that references (links to) articles.

It should not duplicate their (full) texts or show excerpts. That way, you introduce a duplicate content issue, although I believe Google is smart enough to deal with it. Still, it is a problem that can be prevented.

If you change the template of the archive page into a 'reference' (containing only links to articles, not excerpts) instead of a 'duplication', you fix another one of the problems that many publisher / blog websites display in their pagination systems: duplicate content.

There is also another advantage to this way of templating, but first I have to tell you more about an old SEO-technique.

2. Topic pages and the 'third level push'

Now if you have a very large publisher website (like a blog, magazine or newspaper), you can start thinking about the use of 'topic pages'. These are dossiers about a specific subject, for instance about a famous person. The page could inhabit a short biography, a picture, maybe a knowledge graph with some data, the latest and most popular articles about the subject, etc. And of course, also a way to navigate easily to all articles about the subject!

All articles that are connected to this topic page (tagging, anyone?) reference the topic page. That is a lot of related linkjuice that points to the topic page. We call this the 'third level push': your homepage is level one, topic pages level two, and articles are level three. Usually it's your homepage and the articles that receive most inbound links, and thus level three has a lot of linkjuice to give. This is an easy and relevant way to link: from an article you link to the topic page, where people can read more, and see other articles about the same topic. And usually, people search more for the generic name of a topic than for a very specific article name.

In my experience, tag and / or category pages are quite easily turned into topic pages. Most publishing systems already have a tagging and/or category system in place. It's just a question of expanding the template of these pages into a topic page (and replacing the pagination system behind it with a chronological archive).

Problem: QDF

Up until 2011 this technique of topic pages worked really well. But then, Google modified its algorithm ('QDF': Query Deserves Freshness), in order to better accommodate searcher intent when they were looking for a buzzing news topic. It makes no sense to show old news or even well-ranking topic pages when something is 'hot' in the news.

The consequence was, that topic pages didn't work that well anymore. Well, not completely. They do still work very well, but not when the topic is in the news. The rest of the year however, it still works like a charm. Just make sure you have enough of them. And that you know which ones attract traffic. Also, in SEO-circles, many people say that this technique 'does not work anymore', so there is less competition for those who know better ...

Now if you combine topic pages with a chronological archive, then you get ...:

3. Topic pages + chronological archive per topic

Let's just assume your website has a tag or category system available. Keep using it. But instead of pagination, use a chronological archive, per tag. And don't name it 'tag' of 'category' anymore, but something less technical, like a 'dossier', or a 'topic' (that's my personal preference).

Now, all articles that are topically, thematically related, are (and keep being) listed on the same archive page. Google understands the context of these articles better, and human beings can browse through related articles better. Win-win.

And now, the first page of this tagging system should be turned into a topic page. So let's say the URL of a tag called 'Eikhart' (which is a hugely interesting subject of course ;)) would be:

And the URL of all articles written in 2015 about this topic would be:

That's it. The first URL (without a year segment, just the name of the topic) will become a topic page, which contains all kinds of extra interesting stuff about a given topic.

All articles that are related to this topic, will link to it, which makes sure that not only these topic pages are really well interlinked, integrated in your site, but also that they receive a lot of link value / PageRank / link love (whatever term you prefer to describe the value that links give).

Of course, you have to make sure that your topic actually are interesting. Copying Wikipedia texts doesn't cut it.

4. Bonus: news tickers

It's also quite easy to use all these related articles for a news ticker. So let's say that something major happened in Amsterdam: the Dutch football team has won the World Cup and is now celebrating in the Dutch capital. All articles (and I mean ALL of them, also the ones that are 20 years old) that are within the topics 'Amsterdam', 'Netherlands', 'football', 'World Cup' can now display a news ticker that might even refer to the event article itself.

Usually news tickers are either only displayed on the homepage of a newspaper, or on all pages, even on non-related ones. This is an easy way to display it on all related pages instead.

This is just an example of how to be able to expand on this. It starts however, with getting rid of that damned pagination system ;)

What's next?

If you think this was useful to you, it might be for others also. Feel free to share: