Over the years, you've certainly read something about how Google loves fresh content. Perhaps you've read that sometimes it takes its love of freshness too far.
Now it's the middle of 2015. Does freshness still play a significant role in how Google ranks search results?
To find out, I decided to conduct a small experiment on a blog. Specifically, I wanted to see if my test could answer the following questions:
- If you update a blog post's date, will it receive a boost in the search engine results pages (SERPs)?
- Can you fake freshness?
- Do you have to make changes to the content?
- If there is a boost present, how long does it last?
Details of the test
- This test was performed on 16 blog posts on the same site
- All posts were originally published between September 2010 and March 2014. Each post was at least one year old at the time of this experiment.
- Each post (except No. 16) received organic traffic throughout 2014, showing an ability to consistently rank in the SERPs
- URLs for these posts did not change
- The content was not edited at all
- The content of focused on evergreen topics (not the type of queries that would be obvious for Query Deserves Freshness (QDF)
- Only the publishing date was changed. On April 17th, the dates of these posts were set to either April 16th or April 15th, making them all look like they were one to two days old.
- Each blog post shows the publishing date on-page
- Posts were not intentionally shared on social media. A few of the more trafficked posts likely received a couple of tweets/likes/pins, but nothing out of the ordinary.
- Google Search Console, Ahrefs and Open Site Explorer (OSE) did not show any new external links pointed at the posts during the time of testing
Baseline organic traffic
Before starting the test, I took a look at how the test posts were performing in organic search.
The graph below shows the organic traffic received by each of the 16 test posts for the four full weeks (March 15 - April 11) prior to the test beginning.
The important thing to note here is the organic traffic received by each page was relatively static. These posts were not bouncing around, going from 200 visits to 800 visits each week. There is little variation.
The blue line and corresponding number highlights the weekly average for each post, which we will compare to the graph below.
Turning the test on
This one was pretty easy to implement. It took me about 15 minutes to update all of the publishing dates for the blog posts.
All posts were updated on April 17th. I began collecting traffic data again on April 26th, giving Google a week to crawl and process the changes.
Organic traffic after republishing
All 16 posts received a boost in organic traffic.
This graph shows the average organic traffic that each post received for the first four full weeks (April 26 through May 23) after republishing.
I expected a lift, but I was surprised at how significant it was.
Look at some of those posts, doubling in average traffic over a one month period. Crazy.
Faking the date on a blog post had a major impact on my traffic levels.
Post No. 16 received a lift as well, but was too small to register on the graph. The traffic numbers for that post were too low to be statistically significant in any way. It was thrown into the test to see if a post with almost no organic traffic could become relevant entirely from freshness alone.
Percentage lift
The graph below shows the percentage lift each post received in organic traffic.
Post No. 14 above actually received a 663% lift, but it skewed the visibility of the chart data so much that I intentionally cut it off.
The 16 posts received 3,601 organic visits in four weeks, beginning March 15 and ending April 11. (That's an average of 225 organic visits per post, per week.) In the four weeks following republishing, these 16 posts received 6,003 organic visits (an average of 375 organic visits per post, per week).
Overall, there was a 66% lift.
Search impressions (individual post view)
Below you will find a few screenshots from Google Search Console showing the search impressions for a couple of these posts.
Note: Sixteen screenshots seemed like overkill, so here are a few that show a dramatic change. The rest look very similar.
What surprised me the most was how quickly their visibility in the SERPs jumped up.
Keyword rankings
It's safe to assume the lift in search impressions was caused by improved keyword rankings.
I wasn't tracking rankings for all of the queries these posts were targeting, but I was tracking a few.
The first two graphs above show a dramatic improvement in rankings, both going from the middle of the second page to the middle of the first page. The third graph appears to show a smaller boost, but moving a post that is stuck around No. 6 up to the No. 2 spot in Google can lead to a large traffic increase.
Organic traffic (individual posts view)
Here is the weekly organic traffic data for four of the posts in this test.
You can see an annotation in each screenshot below on the week each post was republished. You will notice how relatively flat the traffic is prior to the test, followed by an immediate jump in organic traffic.
These only contain one annotation for the sake of this test, but I recommend that you heavily annotate your analytics accounts when you make website changes.
Why does this work?
Did these posts all receive a major traffic boost just from faking the publishing date alone?
- Better internal linking? Updating a post date brings a post from deep in the archive closer to your blog's home page. Link equity should flow through to it more easily. While that is certainly true, six of the 16 posts above were linked sitewide from the blog sidebar or top navigation. I wouldn't expect those posts to see a dramatic lift from moving up in the feed because they were already well linked from the blog's navigation.
- Mobilegeddon update? In the Search Console screenshots above, you will see the Mobilegeddon update highlighted just a couple of days after the test began. It is clear that each post jumped dramatically before this update hit. The blog that it was tested on had been responsive for over a year, and no other posts saw a dramatic lift during this time period.
- Google loves freshness? I certainly think this is still the case. Old posts that rank well appear to see an immediate boost when their publishing date is updated.
Conclusions
Let's take a second look at the questions I originally hoped this small test would answer:
- If you update a blog post's date, will it receive a boost in the SERPs? Maybe.
- Can you fake freshness? Yes.
- Do you have to make changes to the content? No.
- If there is a boost present, how long does it last? In this case, approximately two months, but you should test!
Should you go update all your post dates?
Go ahead and update a few blog post dates of your own. It's possible you'll see a similar lift in the SERPs. Then report back in a few weeks with the results in the comments on this post.
First, though, remember that the posts used in my test were solid posts that already brought in organic traffic. If your post never ranked to begin with, changing the date isn't going to do much, if anything.
Don't mistake this as a trick for sustained growth or as a significant finding. This is just a small test I ran to satisfy my curiosity. There are a lot of variables that can influence SEO tests, so be sure to run your own tests. Instead of blinding trusting that what you read about working for others on SEO blogs will work for you, draw your own conclusions from your own data.
For now, though, "fresh" content still wins.
Love the experimentation Anthony! I think, in order to confirm the results and to check whether this has a consistent effect, it would be great to spread this to a few dozen blogs and monitor/record/publish the results as you have.
One theory I've got - updating the date made a fresher date appear in the SERPs, which caused CTR from searchers to rise, which caused a rankings and traffic bump. Hard to prove, but it could explain how Google uses more than just raw publication date.
I would love to see some fellow Moz readers conduct a similar test. This was definitely a small test (one site, sixteen posts) so it would be interesting to see how other sites respond.
As for the CTR data improving rank, that could definitely contribute. With some of the posts that were sitting on page 2 for key terms, I wonder if they could receive enough impressions to result in a CTR improvement?
Perhaps the "freshness" makes Google consider testing it higher in the SERPs and the date/CTR improvement helps keep it there.
Hi Anthony,
Great experiment, it's great to get answers to these questions about what "The Google" loves.
I recently added a group of blog posts to a site, and saw large jumps in visibility, including having one of the blog posts appear 5 pages above where the home page for the site was for a long tail (not targeted) keyword.
This lasted for an initial burst of 24-48 hours, then the site as a whole shifted up to it's new normal, and some of those new blogs settled back further. I haven't done it as a full experiment, but rather made note of a few of the terms and the effect of the onsite changes.
If these were already live and had reasonable traffic it would have been very interesting to test what impact it would have and the length of time that impact would have listed.
I'm curious, your blog post with negligible traffic... how much extra traffic did it gain, and has the effect lasted as long as the others?
Thanks!
Post 16 went from 2-3 visits per week to about 5. I don't have the exact data available to me at the moment
Thanks for the reply Anthony, statistically thats still significant.
Hey Anthony, great case study! Just a question though- how big of a difference does it make if you don't show the date on the page?
That was my take on it.
I think the CTR impact could be the ultimate cause. Anecdotally: I had a popular informational blog with evergreen content, and boosted organic traffic by 10% just by removing dates entirely -- under the theory that since Google was showing dates in the SERPs people skipped over my posts for more recent ones. Even posts ranking #1 saw the traffic increase, though I did not look to see whether my rankings improved for lower ranking posts.
Totally agree with the CTR theory. This article should show the titles of the articles in the test to give us a better idea.
I know that I tend to ignore articles which are a few years old, especially if it's SEO-related.
We were following this approach in my last company (3 years back) and I've always seen this technique working in our favor. It was an education base site so every year we just use to update the dates in the last year post (from 2010 to 2011 and 2013) and instead of specifying the published date on the page we have the last updated date. Google always used to show the updated date in the search results.
We wrote an article in 2010 and every following year we've just updated it with the latest dates and specified the year in URLs, trust me this site was generating tons of traffic because of this approach.
At that time we were ranking for some of the highly competitive keywords like Gre 2013, GRE Exam, GMAT, Sat exam, MBA colleges in India and many others test paper related queries.
One last tip: If possible specify the year URLs, it will really make a difference in CTR (in our case it was possible because our's was an education site)
Really interesting experiment, probably CTR was the reason.
I agree with you Rand (CTR).
I did some experiments myself and by adding "updated 2015" The CTR of my posts went up like crazy...and so did my rankings.
I was able to rank 2 posts for the KWs "aprender SEO" & "Aprender Adwords" (learn SEO & learn adwords in Spanish) in the first position for the last 2 years.
Thank for the experiment!
Anthony,
Thanks for sharing this post. I suspect lots of folks will be trying this tactic very shortly. Hopefully, everyone read the part about how the posts in the test were rather meaty to begin with and were receiving decent traffic. Freshening up low quality posts isn't going to help a whole lot.
RS
Hi Ronell,
You're absolutely right that these were all quality posts that have been bringing in organic traffic for quite some time.
While I found this study fascinating, I hope it doesn't just lead to people updating their post dates to potentially see a short traffic boost. Instead, they should look for old posts that can be improved by actually updating the content and republish those newly improved pieces.
Improving existing content can be highly beneficial, as opposed to a little trick or temporary boost from updating a publish date
I was just about to jump and say the exact same thing Anthony!
I will definitely give it a try :)
This is where SEO gets itself in trouble. Before we ask ourselves if we can do something, we should be asking ourself if we really should.
The fact that we're blatantly saying we're "faking" the freshness of a post already should be a warning sign that it's wrong, much more the fact that the community is reading this and considering it as a "best practice" or yet another "trick" in a long line of "hacks" that we just shouldn't be touching.
Instead of faking freshness - just write something new.
Hi Jeff - I think the post and comments are clear that this is simply a trick/test and won't add any long term value. It's just an attempt to learn a bit more about what shows up in the SERPs and have a better understanding of the potential value of republishing old content (hopefully after you update and improve it).
I agree completely and follow that approach to get better numbers. If you are going to go through the trouble of opening an old blog post and redating it, taking an extra 15 minutes to scan it, add a few sentences or paragraphs or make other improvements will enhance it even more and potentially improve its rankings over the long term.
Besides, I have to think that the search engines will eventually catch on to the practice of simply redating and end it.
I've been going through a website that has some seriously old blog posts (2006+) that generate 5-10,000 hits a month. I've been doing a little bit of light editing, and changing the publishing date. The results I've seen have been right in line with your test, huge immediate jump in traffic on the order of 50% more. Granted that I have been doing some cleanup on the posts and comments, but I feel like the publish date has played a larger roll in the increase. I wish I could credit my writing skills with a 50% jump in traffic!
Thanks for sharing.
Great experiment, Anthony!
I'm honestly surprised it would have that big of an effect simply to change the publishing date. Sure, the sample size is small, but it's still a bigger increase than I would've guessed.
I think this is just another incentive to update some of those good old evergreen blog posts, if there weren't enough already. Content creation doesn't always have to be new content.
You are exactly right. I was attempting to isolate the potential value of simply updating the publish date. I don't think that is a viable tactic to increase traffic, but if it can help people consider revisiting their old content and making it better, that is a win.
Nothing new here. Ever since Google cranked up the dial on Query Deserves Freshness - QDF - ANY page can be usually be boosted with an updated time-stamp. Particularly useful for Lists, Link/Resource pages that get updated periodically.
It works on static pages too - not just blog posts. Just add an italicized or bolded date reference like: "Page Last Updated : October 1st, 2015" anywhere in the main body content. Since numeric nn/nn/nn(nn) date format conventions can vary geographically it's best to fully spell-out the month, day, year. It's smart to include words like 'updated' 'published' 'revised' to put the date in context.
Even a "Copyright 2015 : MyCompany" down in the footer is wise to do immediately after the first of the year - It's often an easy way site-wide to send search engines a clear signal that you've got active skin in the game.
Anthony thank you for sharing the results of your efforts here. I agree that the real takeaway isn't to update your publish dates, but to to find good posts that need to be updated and refreshed a little. This should be part of every content audit (keep as-is, remove, or improve) and is much more cost effective than writing new content from scratch. You can offset your content production expenses by improving/updating an old post that has performed well over the years.
Exactly. This test isolated the date only aspect of content freshness. Hopefully it gives people motivation to improve their older content.
This post on the Buffer blog yesterday is a timely companion read for people just finishing up my post:
That is the post I was actually looking for some time, Great experiment Anthony. I have already tried this method back in 2014. One of the post at my blog was performing very well in 2013 and with the passage of time it lost the impressions in search and the traffic too. But what I did was a bit different.:
The post received 400% boost after 2 weeks and up till now this post is getting some good amount of traffic which has not dropped till now despite the fact that we are in 2015 now and the post is more about 2014's trends. I would say the content freshness can be faked to some extent but my experiment was a bit more extensive then just a change of date. Waiting for someone else to post their results now...
Nice. That's a bit different though. Sounds like you made the original post better and more relevant
Wow! I can't believe it is so simple to hack the SERPs :)
Thanks!!
Hey Anthony,
This is an awesome case-study. I always thought doing this could be beneficial (for good posts) and now I have confirmation :D.
I periodically update my articles with fresh information, images, reformat the text and link them to new & relevant articles on the site. At first I started applying this method on top-performing posts, now I do it with 2-3 articles a month just to make sure that older content isn't becoming irrelevant.
However, I never changed the publishing date. I simply added an "Updated on" note at the beginning of the article. I have to test out your method.
Thanks!
This is really good info on a loophole that I can imagine Google in a year or two - maybe sooner if many sites suddenly start doing this to game the system!
I'd love to know how blog posts Without dates compares with these posts with Old dates vs New Dates.
Great post Anthony. Worrying a less-than-honest way to so easily game the SERPs still exists.
Hi Anthony,
I was thinking a lot about Google and content freshness, but your experiment gave me a few answers. Thanks for the insights very interesting results I will definitely give it a try.
Hi Tomzur,
With Google making so many changes, we always need to remember to challenge and test our own assumptions. Sometimes we think "Google is too smart, that won't work anymore" but then two years later, it's still working.
Great to see Anthony, you have published the result, I have done this last month and results are same more or less.
the post were dead in SERPs gets good hits than the other, I would say it was due to the trend of user as they may have been looking for the information was on those pages. The post those were getting traffic on first hand has not changed much.
Initially it was with 20 post after that i have done few more but on different site. will update you guys here as soon as i see the results on later updated post. This time i also has changed the meta description and title too.
That's exactly what I'm doing right now, although I started with one republished post per day and one new post per day. The plan is to go like this for one month and then share the results.
I'm sure that social sharing played a very important role in the freshness factor.
Search engines love new content. That’s usually what we mean when we say ‘fresh’.
So you can’t update your pages (or the publish date) every day thinking that will make them ‘fresh’ and more likely to rank. Nor can you just add new pages constantly, just for the sake of having new pages, and think that gives you a freshness boost.
However, Google does have something it calls “Query Deserved Freshness (QDF)”. If there’s a search that is suddenly very popular versus its normal activity, Google will apply QDF to that term and look to see if there’s any fresh content on that topic. If there is, that new or fresh content is given a boost in search results.(ref: search engine land)
Hi Anthony,
This was a very nice experiment.
Perhaps on phase two you could do something like simply adding a fresh date to the post while leaving the actual date alone. Along with updating sitemaps I wonder how this would impact some older post.
Example
Originally Posted: (January, 4th 2013)
Last Update: (September, 10th, 2015) --Maybe switch these 2 dates around
Fluffy bunny neckties, are making a huge comeback
Yada, yada.............
Once again, this was a nice experiment. I would love to see others results as well.
Thank you,
Don
Thanks for the comment Don. I do not currently plan on a phase two of this experiment.
Exactly what I was thinking. Plus, if you add one sentence to the text in post and then just add "last update" on post and update sitemap, it would be even better, maybe.
Nice Post. Interesting test. Just wondering did you checked the traffic lift considering the visibility of posts on blog's home page feed? How many posts were showing up on your Blog's Homepage feed when you updated the dates? All of 16 or few? As showing up in Home Page feed can be a valid reason for a traffic lift as compare to those posts which are not showing up in the feed. Just saying...
Posts 1-7 made it on to the blog homepage feed. Posts 8-16 were on the 2nd page. Surprisingly, those posts on page two received a very big lift. All traffic data is organic traffic, not a measurement of pageviews.
Great stuff, thanks for sharing! I haven't seen this level of analysis before on fake blogs and I see more of them popping up.
Amazing work Anthony. Would be great if IMEC Lab can help us understand this better by experimenting this on a larger set of blogs (with more number of articles).
While I agree this small test isn't conclusive in any way, I don't think having multiple people or sites verify this is that important. It's not a tactic for sustained growth. What it should be though is extra motivation for improving older content. If simply republishing can lead to increased attention, it should help people build the case to go back and update existing content.
Fantastic test! I've wondered about that myself. I would love to know if adding/using the updated post option in addition to the published date would help just as much. Or if the published date overrides no matter what.
Great article Anthony! One question though, if you don't mind me asking, what type of posts where they? What was the industry, topic, etc?
I'd be interested in this too. There are plenty of searches where I view only recent posts (from the past year as at the most) - particularly when searching SEO topics. This could explain more impressions and lead to a higher CTR as Rand suggested above?
Effective Post , Like it!!!
Interesting case study, Anthony! It's hard to believe if everybody can actually fake freshness to boost performances but as Rand pointed out the experiment should be conducted at a much wider scale to establish whether it works in every single case. I'll run the experiment on a couple of blogs of mine and see how it turns out. Thanks!
I don't think anything in SEO works in every single case. Right now, faking a publish date appears to help boost older content's search visibility. Let us know if your experiment shows similar results.
This is timely for me because I have some great evergreen posts that used to get a lot of traffic but have since been buried by newer posts and I was wondering how I could get them closer to the "top of the list" on my WP site.
I have to ask: would Google consider this manipulative (aka black hat)?
If you are simply faking it like this experiment and abusing it (doing it frequently), maybe. Instead, focus on updating these posts and republishing them once a year. There is nothing manipulative about improving your dated content, the web is a better place when that is done.
Fascinating test, looks pretty conclusive to me! Could it be that updating the date counts simply as editing the text? It would be interesting to compare with editing a few words in the body
Innovative experiment! @Anthony - how would you deal with if the comment of the post is dated earlier than the post published? Do you think it might be an impact on traffic as well?
When republishing anything for real, I would definitely put a "This post was originally published on .... and was last updated on ...." type of message on the bottom.
Anthony,
Great Experiment done. I would Surly give a Try and Change the date of my Blogs. Let's see if that Works for me too.
Thanks for sharing.
Thought of updating old post regularly,is good seo practice.. But my question is that what if we change, and replace full content, and update it. Will it affect the search engine ranking or seo anyhow?
Changing the content on a page will likely change the search engine ranking. Just focus on making your page better (for users & engines).
Nice thinking and its working but I want to know that its will work at upcoming times or not because Google is now become smarter.
This is not a long term tactic to increase traffic. Just a small test to see if it is still working. Hopefully the small test will incentivize people to update and improve their older content.
Incredible post Anthony D. Nelson, it is the most unique and probably the easiest way I have ever read to boost the traffic in your sites. I am quite excited to do this experiment in my few years old articles and will definitely sahre the feedback it it goes perfect! Thanks for your efforts
I think these results are valuable - why not give your content a boost every now and then? Even better, why not update your posts to make sure they are most relevant to the searcher at that point in time. I'm sure that's an efficient way to build an audience.
Genial post, it's excited all the content, I try in my cooking blog with the post are submit, I comment my results as soon as posible. Thanks for this post.
Awesome post. Really enjoyed reading it and am quite shocked that you can get away with this. Interesting to see google doesn't have a way to filter out this sort of faked freshness content.
I read this post and I am performing the test myself on my website ( www.autotransportquoteservices.com ). I only changed the dates of a few pages to test out how well they might jump. I might make this something I do every few weeks to other pages and have it be a continuous event. We shall see...
Loved this case study so much we carried out our own with some surprising results.
To see our results just take a look at The Whole Caboodle - Content Refresh Case Study Results.
Thank you again Anthony D. Nelson
Loved this case study so much we carried out our own with some surprising results.
To see our results just take a look at The Whole Caboodle blog
Thank you again Anthony D.Nelson
Nice post! Time to change the dates for my blog ;-) (just kidding)
Don't publish the work of the devil. Too many people have been doing this for years! It drives me crazy when I try to filter Google searches to the last year etc. and I get stuff that is 3-5 years old. It might be really hard for Google to figure this out, when they do I wish they would "permanently" blacklist the hosting domains.
Wow, that's a nice writeup Anthony. I would expect more boost if the content is also updated accordingly. What do you say?
Absolutely agree that improving the content is much more important. This was just a small study to isolate if "freshness" can be faked and what type of impact it can have.
Great post.this is really amazing post!!
Did you updated dates in sitemap as well?
No. There were no dates in the sitemap.
Great information, bookmarked for later use. Thanks for your help. #1 thumbs up.
Its always interesting to see the different ways you can write content and rank on google search engines. I still feel however that Quality content and organic SEO will always prevail
Thanks for sharing
Love the information AND testing....will do the same--THANKS SOO much!
Hi Ronell,
I think it's important that people realise that you can't just update your post dates in the hope to see a short traffic boost. The best way to do this would be to find your old content, edit and improve it and then republish them. Many forget about improving existing content when the results can be highly effective.
Thank you for the great article,
Ben - Co-owner of PurpleFruit Digital Marketing Agency
great post Antony will try this for my blog post
Hi you cheeky bastards, you got me thinking on Friday ahead of weekend ;)
Here we are - Posts have a publish date clearly visible in google's snippet, pages don't.
How do you "refresh" publish date for pages then? Does changing the publish date for pages (let's say in wordpress) make any changes for ggl as posts do?
One of the best youmoz posts I read in last few month - maybe just because I simply love all kinds of tests. I tried that in an wordpress blog (early 2013) in a "no-freshness-needed"-topic (thats what I thought) and the effect on a lot of posts was an equal lift of visitors. But I think more posts, more blogs, more topics are needed. Just use 5-6 posts is (like u said) bring them closer to the start page...
Hey Anthony, thanks for the post & happy to see you experimented something small but which resulted in higher CTR & traffic to your blog post.
Still Google is lacking advanced mechanism to tackle this solution. I hope after reading your post they might come up with some. I remember, Jason Acidre of Kaiserthesage tested the same thing but with different concept. He found in analytics that people are visiting his blog post with some different keywords & so he added those keywords in the blog post & that boosted his traffic again.
I really wish to promote this post so that more users can test this.
Thanks.
I remember that post of Jason's. Definitely want to reiterate that the real value here is in motivating people to go back and update and improve older content. This test showed that simply updating the date in isolation can have a positive impact, which alone should encourage people to revisit their old content.
I'm glad that was the spirit/intent of the post at least; however, knowing the SEO community, that's not how it will be used.
This post really effective. thanks for share.
Nice experiment Anthony. There are many blog sites where we can change original publishing date. I did it with many of my blog posts. Just to hide original date from clients. I am not able to post any comment about experiment because i have to check results. See you.....
Hello Anthony,
Well done mate, its definitely a smart way to serve the same food with different gravy ;) . I really enjoyed your post, let me try it quickly for my own post, as i can't wait to see the boost.
Such kind of posts make MOZ a better platform to learn smart, interesting & unique things about our industry..
Thanks and keep sharing such an amazing stuffs Anthony :)
Such a lovely experiment Anthony!! results are way more than expectation i would say.
I was surprised as well. When I ran the test, I didn't think it would result in anything publish worthy.
Hiii Anthony, Great Experiment I read it very carefully at every step. But it could be work if I don,t display dates in my post.
The test site had the dates displayed on the page, but not in the URL. Would be interesting to see someone test it on a site with no dates.
Here are some interesting links on refreshing content:
SERPS
SearchEngineLand
Great inititiative!
Besides the clear benefits of altering the publication date, I think it's always good to check your existing content and regularly update it. Updating content to reflect the latest insights on a specific topic will make it futureproof. It helps you to make and maintain that your blogs content is authorative.
On topic: A plugin that automatically changes your publication date once every few months would be excellent to use.
The plugin idea is excellent. It could also be coupled with certain categories or tags, and there you have it: an automated, article laundering system. I'll add the idea on my list, maybe I'll code something this weekend.
Thanks, Thomas!
Where's the data for the control set?
Wow - didn't expect such a big boost to the results. Am off to try my own experiment
Please come back and share your results in the comments.
I really like you post thats why I also shared it on social media.
Thanks for sharing it Rahul.
Great post. I have often wondered this but have never gone to this effort. Thanks for another tool/toy to add to my kit bag!!
Hi Anthony,
What if I completely remove all dates from my blog? As I couldn't able to write articles frequently.
Does it help me to get good organic traffic?
What is your opinion?
Thanks
Fal
I would not recommend this. If you can't write frequently, just put your focus on quality. Aim for one great piece of content a quarter or whatever works for you.
Good experiment !!!
Great experiment, but I would echo those that say other factors would also be at play. It's probably also worth pointing out that this is not a good idea long term - it wouldn't surprise me if Google keeps records of original publication dates and subsequent published dates. If you create a pattern, they will spot it, especially if the content hasn't changed (which it didn't in this test).
Evergreen Content should be periodically updated and then it's okay to amend the publication dates, just be careful not to think that this is a quick "SEO Win".
I agree with everything you've said here.
Expanding on Donford & Martin Oxby's points,
I strongly believe if a website has enough organic traffic to conduct a test like this, then Google doesn't need simple markup such as dateCreated, dateModified, datePublished because it's algorithm(s) / index identify those characteristics very effectively. Assuming that is true, then Google does not believe this content is newly published.
This is without considering that links, comments, and page structure all build a timeline of information about a page.
My theories on the traffic bump would start with site credibility, site structure, the small modification, and article position.
Thank you for putting in the work and sharing the results Anthony!
Thanks for this post. It is surprising (worrying?) to see how much of a difference changing the date makes.
Did you mean 3,601 and 6,003 are the average per week in the following paragraph otherwise the maths don't seem to add up?
"The 16 posts received 3,601 organic visits in four weeks, beginning March 15 and ending April 11. (That's an average of 225 organic visits per post, per week.) In the four weeks following republishing, these 16 posts received 6,003 organic visits (an average of 375 organic visits per post, per week)."
3,601 and 6,003 are in fact the weekly averages for the four weeks pre/post date change.
This post shows as live on 9/9/2015 (2 days ago), but some of the comments are noted as written a month ago... Did I miss this post on the last go-around, and now you've updated the date?
It was published last month on YouMoz. When a popular post gets promoted to the main Moz blog, it receives a new publication date.
Great thing in studies, are that you can get some info faster than doing study by yourself. They are always welcome! Thanks for sharing.
I was not expecting such results from Google. Anyhow thanks Anthony for sharing your case study with us. It’s really interesting.
This question was haunting on my mind since long time. After reading your case studies, i am sure, lots of people will start doing experimenting on it and soon you will see (may be or may not be) the consequences of this post practically. Regarding Content Freshness, this is the only article which is based on case studies with proven results. Enjoyed it, very very effectively written post.. thanks @Anthony.
So I'm guessing this was one of the test posts for Moz :) I had to do a double take when I saw the "published" date and the comment dates, but hopefully this will yield some interesting data. I did not find this through organic search; I found this post through Feedly. Hopefully we'll see some data from Moz's test from "faking freshness" in the near future.
Anthony - Great post btw. I hope you can share more tests like this in the future, it was an awesome read!
This post was published on YouMoz last month and now promoted to the main blog, therefore republished.
Awesome experiment, thanks for the share! I think I'll start devoting more time to updating old blog posts from now on.
Did you check the terms you were ranking for before vs after?
I ask, because, I think adding 2015 to the posts has a lot to do with it, not only because of freshness, but because 2015 in on the page.
You probably started ranking for a lot more terms like ["search term" 2015]
I think a lot of people add the year to search terms so they get fresh results, especially after they search with just the desired term without date and get many old blog posts.
All of these posts were Evergreen content. Think "how to change you lawnmower's oil" for an example. I wouldn't expect any of the posts to have people searching for the year with the keyword query.
Feel free to delete this comment, I posted it under the wrong login.
Hi Anthony,
This is a very interesting article! Thank you for taking the time to do something like this. Here are some questions I have for you:
All in all, this was awesome work! Congratulations!
Posts #1-7 above hit the first page of the blog. Posts 8-16 would be on the first archive page. The "real" posts were not currently receiving any organic traffic, so their traffic levels were unchanged. I find that actual fresh content doesn't rank that well for competitive terms, likely because there are no external links pointed at the pages.
Nice post. Thank you for sharing!!