Republishing older posts: pros and cons
These are the stated benefits from recycling content:
- Attract extra traffic for little extra work. While we have updated and edited these posts, it is less work then creating new content.
- Attract new backlinks. While some of these older posts had good links, over time these can be broken and republishing can help to add new links.
- Resurface content that is useful for our audience. We have a good back catalogue on Econsultancy but, thanks in part to less than optimal site search and navigation, some good posts are hard to find.
- Create more time to produce fresh content. Using older content helps us to keep traffic flowing while we work on newer articles.
- Create new leads. These older posts were linked to newer related reports, training courses and events, allowing us to create leads.
My concerns about reusing content:
- It’s ‘cheating’. This is a personal thing, but it does feel a little too easy to do and, as someone who believe in creating original quality content it doesn’t always sit well.
- Some may view it as misleading. We have labelled the articles as republished, updated versions but it’s possible some may miss the notes. I don’t think this is a serious issue, as the content is useful and, let’s face it, we wrote it in the first place.
- It messes with the order of the blog. Everything (at least until we started this experiment) is there in chronological order on the blog, so you can see the development of this blog from the very first post nine years ago, to today. In truth, that probably only bothers me.
How we repurposed old posts
The Hubspot article contains some useful tips on this, and I won’t repeat them here. Here’s what we did though.
Select some candidates for repurposing
I decided that we’d confine this to posts that were at least a year old, to ensure that they’re not too fresh in the memory.
Using Google Analytics, we looked at some older posts which were still relatively popular, but not too popular.
For example, seven of the ten posts here are older than a year (only 2, 3 and 9 aren’t) and are still doing very well.
They’re good examples of evergreen content and don’t need any extra help from republishing.
So we scrolled a little further down the list, looking for the good stuff that doesn’t bring in quite the same amount of traffic (these stats show roughly 12 months of sessions). Maybe posts bringing in respectable but not amazing numbers (we weren’t that scientific).
Others, for obvious reasons, don’t merit republishing. The cookie law is less relevant now as it was then, when people were keen to see how sites were implementing cookie notices. Other posts were tied to an event, such as the ’25 years of the web’ article.
The one I chose here is, I think, still relevant to our audience, while I also thought it was an article which didn’t originally perform quite as well as it should have. Here it is.
We chose another ten or so articles, based on the same criteria.
Before we hit the republish button, we recorded some key stats so we could measure the results.
Unfortunately, as the site has been redesigned since these posts were originally published, the social shares counts are out, though we can piece this together using other tools.
We were primarily interested in:
- Backlinks. Using Majestic, we noted the number of backlinks prior to reposting. The backlink numbers generally ranged from 60 or so to 200, though this one had 13,000 backlinks (I need to find the secret and bottle it).
- Pageviews. How many new views would these posts attract? As they were, theoretically anyway, proven posts, they ought to do well.
Anyway, a Google doc records these numbers.
We added a note at the foot of the posts to advise that it was an updated and republished version of an older post.
We don’t want to deceive people after all. This was also the approach used by Hubspot, and I think it’s a reasonable one. We also asked guest writers if they were happy with us refreshing their older posts.
I don’t believe we are deceiving people, merely resurfacing older content which the audience will either appreciate or not. Besides, in the case of posts like that on ecommerce product pages, an update is useful.
When we first pressed unpublish and then republish on a post, we received this error message. I thought we’d broken the blog:
For some reason, our backend can’t handle changing of publish dates so well. It works, but the error message appears anyway.
We did wonder whether or not to delete older comments as this does date posts, though for reasons of transparency, and the fact that many of these comments are very insightful we decided it was best to keep them.
For the Ecommerce product page article I highlighted in the earlier screenshot, we originally had:
- 33,421 pageviews.
- 2 backlinks. Majestic does show there had been seven links over the last five years, but five had obviously been broken.
Publishing anew brought another 30 new links, as the Majestic data shows:
It also brought in 7,898 new pageviews, which is a good amount. For context, this would put it in the top 10 new posts on most months.
To add to this, I can see from Google Analytics that it sent traffic to the other posts and pages linked to in the article. I’d estimate it sent another 800-1,000 views to related Econsultancy content.
It wasn’t as popular as when it was very first published, as we can see here, but I’m happy with the results.
This wasn’t necessarily the best result either. One article attracted 1,000+ additional backlinks and an extra 12,000 pageviews.
The other results
I don’t want to reveal all of our traffic secrets, but a ballpark figure from ten refreshed posts would be:
- 2,300 backlinks generated (according to Majestic).
- 56,000+ pageviews.
To add to this, the posts generated traffic to related content through links in the articles and content recommendations.
Leads are harder to pin down, as we tend to move customers offline to sell subscriptions. I can, however, see that these posts generated traffic to our paid content.
In terms of links and traffic generated for the effort expended it certainly worked for us.
Refreshing a post took perhaps 15 to 30 minutes on average compared to maybe three to four hours to create a detailed post from scratch, so it’s definitely an efficient use of time.
However, I’m still undecided on whether to continue the tactic, or perhaps to reduce the frequency. One thing is that we can’t do this forever, as we have a finite number of posts to repurpose.
I’d like your opinion here – do you see this as a valid and valuable tactic? Do you feel that we’re cheating in any way? Is it useful to read updated posts? Would you like us to stop it?
I’m looking forward to reading your thoughts on this…