Pruning Low-Quality and Outdated Content

[ergophobe]

Have you ever pruned low-quality content or outdated content from your site?

I'm looking at a site now that has a lot of content that is perfectly respectable in its way - it's original and some of it was once quite good - but many of these pages now have very high bounce rates, low time on page, low CTR from Google search results.

Examples run the gamut from pages that had their day to pages that never did.

At one extreme: a page that was once highly ranked and saw thousands of unique pageviews per month, generated dozens of comments, but it concerns solving a problem for a version of software that was launched in 2007 and superseded in 2008. So in other words a really high-quality, detailed how-to, but now it ranks poorly and has a high bounce rate and low time on page.

At the other extreme would be a page that poses a question and just quotes the best answer available on a forum from 2009. Never had great value.

Other examples would include many pages about apps, server software and so on that are now quite outdated. It appears that when people do click through, they see the date and just bounce.

Yes, I could change the dates or hide them, but that is not actually my goal. I have no interest in using some level of deception to boost the numbers for the sake of boosting the numbers. What I'm interested in is improving the signal to noise ratio.

Have you culled content to focus results?

From a practical, content management point of view, how did you manage your culling?

What did you do with the pages you culled - take them offline and let them go 404? Change them to noindex but leave them on the site?

[tangor]

For sites which are info related (tech, literature, what ever it might be) outdated materials were moved to an archive location on the site, MARKED ARCHIVE per page, and left alone, with a link, if appropriate, to newer/similar content on the active side.

On other sites I just kill the info if outdated and has no archive value and return 410 for those for at least two years.

[ergophobe]

Thanks. Yes, I'm thinking of the first case - info sites.

So, just to drill down the specifics of how you do it
- the URL stays the same, but administratively they go in the archive section, or do you literally have example.com/archive/page1?
- do you just leave them alone otherwise or do you noindex or anything like that?

[tangor]

I physically move the files ... and that makes it easier for me to identify content at the folder level and the internal url(s) is a simple search/replace. Also, I avoid the potential of duplicating a filename in the future (which could cause different kinds of problems).

[ergophobe]

>>physically

This always makes me sort of laugh in a digital context (I don't mean any disrespect, I just mean my mind starts thinking of grabbing electrons and moving them around the file allocation table).

So then

~/public_html/page1

becomes

~/public_html//archive/page1

and barring an internal link from another page, it becomes unreachable through the site nav, unreachable through crawl, not on a sitemap.xml, but it stays live on an info site and is not flagged as noindex.

Am I describing that right?

[tangor]

Correct. (chuckling at the grabbing electrons)

[ergophobe]

That sounds like a reasonable approach. Thanks for sharing

[EditorialGuy]

Some things I update, some I mark "archived," and some I delete (depending on traffic, time required for maintenance, and whether the pages are relevant to our main topics or are random pages that once served a function but now seem unnecessary).

[koan]

Good on you for not hiding the date, that is probably one of my worst pet peeve when searching for solutions on tech / programming / software issues, where the version and date is always crucial, I bounce right back from any sites that hide the publication date on the article because I see it as disingenuous. On the other hand, that's one of the reason why i stay away from publishing tech content myself, as much as I like the topic, the shelf life is too short.

[ergophobe]

What criteria do you use for deciding what to archive and what to delete?

How has it affected traffic or engagement (comments, form submissions, however you measure it)?

Since posting this question, I came across a podcast where the person said that he deleted 1/3 of his content and saw traffic triple. Anybody seen results like that after a major content audit?

What are you favorite tools? Are you looking at bounce rate? Site entrances?

PS - heading out camping for the weekend, so I probably wont' reply again until Monday... very interested in hearing more nitty gritty on how people decide which babies to kill.

[JS_Harris]

I deleted 2/3 of my content and saw traffic double but I don't buy the idea that the content being gone was the sole trigger for the traffic increase. My revised internal link structure featured the 'prized' content more heavily and that is what made it rise in rank and traffic, in my opinion. Each page has a pair of navigational links vs the dozen or so it had before and each article is better interlinked than it was before.

Take the opportunity to plan navigational changes while you improve content, both are equally important.

[keyplyr]

I will never again publish content that relies on updated currency to remain relevant.

I have an event guide divided into 12 months listing hundreds of international niche events. While this is a useful service which garners good traffic, the maintenance has become a burden. However, this type of info has little value if archived & removing these pages entirely would fail the user.

[EditorialGuy]

What criteria do you use for deciding what to archive and what to delete?

In our case, it's what I said earlier: the decision is based on (1) traffic, (2) time required for maintenance, and (3) whether the pages are relevant to our main topics or are random pages that once served a function but now seem unnecessary.

The importance of (3) shouldn't be underestimated. Sites evolve over time, and so do content or publication strategies.

When our current site was first launched as two previous sites in 1997, it had similarities to a magazine, with new articles each week on a wide variety of subtopics under our broad main topic. (Many blogs today have that same publishing approach.)

Over the years, our site has evolved to the point where it's largely coalesced around a half-dozen or so major subtopics, so a lot of those old evergreen articles on this or that could be termed "editorial outliers" on the site of today

Just as important, other niche sites have come along to cover those "editorial outlier" subtopics much better than we do, so there's no longer a compelling reason for us to maintain articles on minor subtopics that don't especially interest us or most of our readers.

Since we aren't supposed to use real-life examples here, let me use a made-up example that has nothing to do with us: baked goods.

Imagine that, in 1997, we had a lot of content about doughnuts and muffins. We were still in a magazine mindset, so we wanted to do something new every week. Because there weren't a whole lot of Web sites about breakfast sausage and bacon, we published a number of small articles about those and other breakfast topics while continuing to build our doughnuts and muffins sections into major, authoritative sites. Nowadays, most of our traffic is on our doughnuts and muffins sections, and other people's dedicated sites about breakfast sausage, bacon, coffee, etc. have far more information and insights about those non-doughnut/non-muffin topics than we could ever hope to publish (or would want to publish). So it makes sense for us to get rid of the articles about breakfast sausage, bacon, coffee, etc. and focus on the topics that we know and own: doughnuts and muffins.

[JAB Creations]

Why change a URL? Think of a page's worth like credit, the more crazy stuff you do the worse it looks. I wouldn't change the URL if it's example.com/problem-software-2007/ however if the URL is example.com/problem-software/ and it could conflict with a current or future article then yes. There should be no reason to archive or change it otherwise. The more stable your URLs in the long term the more well thought out and better your execution in general. If you do change a URL you absolutely should have a redirect setup as search engines will definitely still crawl those pages looking for changes. If it's a 2007 specific issue then the crawl rate will be minimal though certainly it will continue. Google crawls different sites subjective to their context. In example when I need less quality but a faster response for code I'll go to Stack Exchange. Often I won't get an answer I need, resume searching on Google and see the URL/entry for the question I just created literally less than two minutes ago. Google is sensitive to time though also to execution.

John

[tangor]

All good thoughts ... but when a page has actually outlived a useful purpose, then removing it (or archiving it in a non-indexed area) will keep site link structure tight and light. Many times "more" is not desired. A SE is more likely to stay up to date with a 1,000-10,000 link site rather than a 100,000-million plus link site.

[EditorialGuy]

I wouldn't dream of archiving pages that deserve archiving in a "non-indexed area." That would make the pages inaccessible to people who might be interested in them.

Someone with a bent for history or nostalgia (or who's researching a topic) might love to read an article about HotDog Pro circa 1995, an S.S. Norway ship review, or an analysis of the 2008 Democratic Presidential campaign that was written halfway through the primary season. Why make such useful archival materials hard for researchers to find?

[tangor]

Thought apparently was not completely stated. Pages that no longer have value should be removed. Those that have value from historic perspective but no longer valid as research are removed from the mainstream ... I keep an archive section which is easily found as the archive area, and titles of those archived pages are listed, but the contents inside are NOT indexed on the big se's, I use an internal search for that content. The archive index page IS indexed by the biggies, the content they point to is not as it is out dated, perhaps inaccurate (due to changes), etc.

All this relates specifically to each niche and criteria. What works for me might not work for another!

[Panthro]

One important thing to consider, I think, is any internal and external links on those pages. By culling a large section of content you are almost unavoidably altering the structure of your site. Any sites linked from those pages will also be affected.

[ergophobe]

Thanks guys. Much food for thought (pun partially intended EditorialGuy).

Many of you mention archiving, internal link structure and such. What's the difference between archiving something in a non-indexed area (I assume you mean rel=noindex on all pages) and simply removing a page from the navigation and all internal links?

Of course I know from a technical point of view what the difference is. What I mean is, over time, with little traffic, diluted link juice and so on and so forth, won't these pages naturally fall from the rankings? And if they don't, perhaps there's a message there.

[martinibuster]

I remove all advertising and orphan the pages I don't have time to fix. When the time arrives I fix them and link them back up.

[piatkow]

I will never again publish content that relies on updated currency to remain relevant.

As the publisher of a music event guide both in print and online that is the day job for me. I use scripts to display an "amber warning" on any information that is at risk of being out of date and a subsequent "red warning" to me to tell me to remove it.

[pshea]

ergo: I would htaccess redirect to a page that represents your current focus. Change all internal links to that page. Then hit the delete key unless the material is really of value. If you are the official source for the hardware or software, then put it in the archives and keep it alive. Like throwing out a high school trophy.