Removing date from articles - Ranking impact?

[serenoo]

My website has about 50 pages. I add one or two pages every week. Every page has at the bottom this sentence:
Last modified on [date]
I would remove this sentence from any article of my website. The date appears on any snapshot of mine on Google searches. I cannot remove the sentence for one page and check what happens cause it is a common proprierty of the CMS, so I have to remove it all at once. Could that impact negatively on my ranking or Google will ignore it?

[Wilburforce]

Google will get the last-modified date from the server whether or not that sentence is displayed, and removing it should not affect ranking. If the information isn't useful to human viewers I would remove it.

[Andy Langton]

Depending on your CMS, there might be other ways for Google to see a publication date. Google also inserts it's own crawl date in certain circumstances.

In terms of rankings, other than 'timely' content like news articles, I've never seen Google attribute much weight to whether something is new or not. Indeed, I have pages ranking with dates over 10 years old on them. Do you want to remove them because you're worried they will get less clicks if they are perceived as being out of date?

[serenoo]

Thank you for the answers. Yes Andy Langton now my guides/articles are ok because they are 2-3 months old, but on 2018 if a user is searching for "how to create the red widget" I do not want that appears 21 May 2015 on the snapshot, because the visitor will find out that it is an old article and I will have less chances to get a click.

[Andy Langton]

I don't actually have any recent data to base this on, but I would guess that it is heavily dependent on your content, and what your audience is looking for.

The data I have does not suggest a large effect, and I see 'expected' clickthrough rates for content with very old dates (from 5 to 10 years ago) across different content types. So, it may be that this less of a problem than you might think. I should probably test it and see what happens!

[netmeg]

I do not want that appears 21 May 2015 on the snapshot, because the visitor will find out that it is an old article and I will have less chances to get a click.

And I and a bunch of other users will HATE you. If I could set my filter to be permanently only the past year, I would do in a heartbeat; instead I have to set it to a year or less for 4/5ths of the search I do. I don't know what your niche is, but there aren't many that are truly evergreen. Removing the dates might get you that extra click, but it also might get you that extra back button when your user isn't happy with the freshness of your post. Just sayin.

[EditorialGuy]

I don't know what your niche is, but there aren't many that are truly evergreen.

It isn't the niche that's evergreen, it's the content that's evergreen--or not, as the case may be. A biography of St. Catherine of Siena is likely to be evergreen; a calendar of 2015 events featuring St. Catherine of Siena is not.

Also, evergreen content can be updated. Some of our most popular and useful articles date back to the 1990s but are updated several times a year.

[Robert Charlton]

Removing the dates might get you that extra click, but it also might get you that extra back button when your user isn't happy with the freshness of your post. Just sayin.

I very much agree... and the more research I do over time and encounter undated material, the more I dislike the trend of removing dates from content that isn't evergreen, and of cheating on dates that Google might use in the serps. I'm surprised that Google hasn't made it a quality issue.

It depends on the nature of content and the niche, of course. Security and technical information, SEO articles, etc, can be very difficult to evaluate with the year removed.

In response to the OP, I do understand the concern... but I'd like to think that anyone searching for 'how to create a red widget' would know whether the material is time dependent.

CMS permitting... for material with a time factor, best way of treating articles that are old but which has been checked for freshness or updated might be to note both the original publication date and an updated date on the article... but only for information you can actually vouch for. I see this happening on sites I trust, and it doesn't seem to have hurt them.

I inherently mistrust sites that automatically post a daily refresh date.

[EditorialGuy]

I often use "Last updated...." datelines on revised evergreen articles, but deciding what date to use can be tricky at times. Let's say, for example, that you've got an article about getting from Newark Airport to Manhattan. The train fare changes, so you quickly update the train fare. That's all you update, because you don't have time to review the entire article (which you last updated a month ago in any case).

- If you change "Last updated..." to the current date, you're implying that everything in the article is accurate as of today (which may or may not be true).

- If you leave "Last updated..." at the previous date, you're telling readers who have heard about the train-fare increase that your article is out of date (which isn't likely to be true).

Things aren't always as black and white as they seem.

[Hoople]

Google will get the last-modified date from the server whether or not that sentence is displayed

On a domain that has a single page G displayed the date portion of "Created on Xxxx xx 2011" as the beginning of the SERP text in grey.

In response to this thread I had removed it and kept only the "Last updated November 25 2015". HTML 4.01 loose, no CMS.

A few days later the date is now gone. Thanks for the motivation to expirement <G>

[Robert Charlton]

Hoople, thanks for sharing that.

-----

And re EG's post...

- If you leave "Last updated..." at the previous date, you're telling readers who have heard about the train-fare increase that your article is out of date (which isn't likely to be true).

Excellent point. Perhaps an update note (added manually, when you edit the article) could simply say something like: "Updated fare information, as of {date}". Doesn't completely spell out all possibilities, but it does help clarify what was done.


PS: Edited formatting for clarity.

[serenoo]

My niche is gambling. I usually have articles with reviews of casinos, casino lists, ... my website is not about news, but about guides, how to win, how to play, rules of the game, ....
I think I will remove the dates, but I was thinking that I will update 55 pages in one second. Are we sure that Google "should" ignore that bulk update?

[tangor]

Are we sure that Google "should" ignore that bulk update?

Are we sure the sky is blue?

Given the above there's no worry attached. None of that data is date sensitive (though I'm not saying G can tell when a poker tournament begins or ends).

Created dates, such as posts in this forum, are important. As for pages general to the web not so much, unless they are NEWS/CURRENT events.

Delete away and no worry.

[alika]

If you are concerned with Google, you can try the approach I did.

I have hidden the dates through CSS from the users, so visitors are not seeing the dates for our how-to guides and other evergreen content. But Google is still showing the dates in the SERPs.

[taryn]

Since I tend to agree with nutmeg in that I prefer dates for content, I use both an original published date and an updated date (where applicable). One, or the other, or both, should satisfy most users. As for google, they have their ways as Wilburforce stated.

[Kirock]

< moved from another location >


Matt Cutts et al have made much of keeping articles fresh by keeping them updated. What I can't seem determine is whether re-dating an old article (that has been updated) seems to matter in the long run.

I've done some limited tests with editing old articles and either re-dating them, or just re-saving them with the original date. It seems that re-dating gets a small initial uptick, perhaps because the pages re-appear on the front page of the site and may get a little traffic due to that. In the long run, however, it doesn't seem to make much of a difference. But my tests have been limited, and I'm wondering if anyone has tested this more extensively?

I wonder also if the freshness factor eventually comes into play if I don't re-date posts that I've updated. Keeping the original date also seems to have some benefit, as readers can see how long you've been discussing a particular topic.

Is there any consensus if you should re-date an old post which has been updated, or just update it and keep the original date? Some of my posts I update every few months as information changes.

[edited by: Robert_Charlton at 7:43 pm (utc) on Jan 2, 2016]

[Robert Charlton]

Kirock, welcome to the forum. I've moved your post from a thread you've submitted, as it's a good fit here, and several points you make are excellent additions to the discussion.

Best practices IMO would be to keep the original date, note updates where dates are relevant (and what's been updated if a partial update only). I also suggest providing references to earlier articles or posts on related topics if they are helpful.

The influence of the freshness factor, I'm thinking, would depend on the nature of the topic, and in particular whether QDF (Query Deserves Freshness) might be a consideration.

[tangor]

If a date is show on any page, post or document, and it has changed by either edit or update, then change the date. The create or modified date will show in other less obvious ways, but if you want your users to know there's been a change, then do so.

As for whether g or any other SE will use that date (or any date other than their access of initial and then see mods/update) I can't say. I would tend to believe it would be all to easy to take last year's widget ad and slap a new date and call it the "new and improved widget".

Use dates for your own internal records and for your users. That said, a date on most pages is not the first thing a user looks for unless it is product, event, or news.

As for dates changing in your specific, they should change when you change the post... and that will return it to the top, once again, just as this web forum does, for example.

[toidi]

If the article is time sensitive and you change only the date, you might fool a se but you will piss of the viewers.

[JS_Harris]

The date only belongs on the page if it helps your visitors, Google doesn't need it. You mentioned using a CMS, its very likely that your date is in the the RSS snippets and elsewhere already.

[Kirock]

Thanks to everyone who replied. I never simply re-date posts to make them look fresh, as I don't think that is good for the reader. I've continued to test both methods (dating, not re-dating - but noting updates either way), and there doesn't seem to be any rhyme or reason to their effect. My new rule of thumb is that if a post is ranking well, don't mess with the publication date (just in case). If it's not performing well, giving it a significant overhaul (if possible) and new date certainly can't hurt, and usually helps.

I'm starting to think that on G's end, the date that they first become aware of a post is more important than whatever the date is shown on my end. I still get crawlers looking for posts that I removed many years ago. I've also noticed that small updates (regardless of re-dating) do virtually nothing in the serps. But if I bump a 500-word article up to 1000, for example, (or combine two small ones into one) I seem to get much better results.

I recently took 8 small/old posts (200ish words each) and made them into one long one (1600ish words) with a new date, and went from nowhere on the radar on any of those 8, straight to #1 in that key word.

[Storiale]

We did a bunch of updates last year - but not all updates are complete yet. We tested updated dates (updated those dates) but kept the word "Update on" below the H1 Tag. We got some uptick in ranking for those How To Guides. We also A/B split test with How To Guides while adding Schema Dates - the results were the same.

The test we have had in our Jira Ticket Cue is to simply comment out the "Updated on" text so that the only date that is left is the last updated date. We refresh each page every 7 days if no update on that page is made and within 6 hours on any page that was updated - I say this because the XML document displays the date in which the page was last updated as well.

I have bothered our programmers to help prioritize the html comment tags around the text "Updated on" text and it is a very low priority for them... In fact they laughed at my urging. (low priority for them). I can only say that I can post results within a week after that template update is made.

I hope this helps...Also - don't forget something.

Date does matter. Links to a page matter, but when all things are somewhat equal, date matters. Date REALLY matters in YouTube now - to the detriment of the site. Run some queries to find a video that was really popular and high quality from a few years ago. The SERPS have been replaced by less popular, less liked and even digital copies of the original video (poor sound) instead of the slightly older and better versions. It is an issue with me and a huge mistake by YouTube/Google.

A good filtering system for YouTube SERPS might not be date, but rather View Count + Thumbs Up Count being 10Xs greater than Thumbs Down Count. I have found this to be a much better calculation for useful videos than just date. IMO.

[timemachined]

I remove date on my sites as just looks untidy and unnecessary. It's reviews and keyword content that is written to last a life time. If however it's an offer and short term, that is included in the on page text somewhere with "if expired, you may wish to check 'page' for other ideas."

I also don't like the date showing in serps. However, on one site I intentionally put the month and year in the actual text for search purposes and search and replace sitewide each month. People buying now, don't want to read about it from ten years ago - IMO. As for outdated content, obviously it's in my best interest to update an article if a line is discontinued, change links or adapt text to refer to something similar to retain possible click through.

As for your question, no I have articles without date on the first page of terms and some of them have been there three to four years.

[paranoid android]

< moved from another location >


We have a number of articles that appear on page 1 on Google for some fairly competitive search terms in our niche.

We've found that if we edit the article and change the date to something more recent, it has a significant impact on traffic. Google usually puts the article back on page one after it has slipped down a couple of places. The spike lasts around 1-2 weeks before levelling off and then eventually fizzing out altogether when the search term is back on page 2 or 3.

My question is, how often should these article be updated? Is it common (acceptable) practise to continually update the article and change the date on it?

[edited by: Robert_Charlton at 4:21 am (utc) on Mar 2, 2016]
[edit reason] Moved post to this thread. [/edit]

[JS_Harris]

- A date is useful to the reader so it might be included in a ranking factor, only Google knows
- A date isn't something all web pages have, or even need, so if it is a ranking factor it's a small one
- A publish date that changes isn't very helpful, it's better to add an 'updated on' secondary date
- A date on the page won't change when Google timestamps what they find, age of page would go by timestamp, not date on page

As others point out Google gets some information in odd ways, actually I should say they record ALL information ALL the time but they use odd information sometimes. Example: They sometimes add whatever name is beside the copyright mark from your footer into the end of your title(s) in serps instead of just the domain URI. I know this because I accidentally had a typo on the copyright once that appeared nowhere else and was used by Google in page titles.

[paranoid android]

All I know is that if i update an article and update the "published date" on the website - the article itself jumps up a few notches in search results.

I'd only do this if I can actually add something additional to the article, and at the bottom i usually put "This article was originally published on xxx, and is updated frequently".

Perhaps Google give different weight to the updated date on an article depending on the site?

[Storiale]

I am still waiting for our programmers to comment-out the word "UPDATED" on our How-To Guides. It's been 6 months. However, I'm confident that rankings will increase a bit.

All our pages are regenerated every 6 hours (if updated) every 7 days when they are not, so sitemaps will indicate fresh date. The date on the Guide is usually within the last year. I'm still hopeful.

[EditorialGuy]

In my opinion, publication dates, "last updated" dates, etc. are too easily gamed to be useful in a general ranking algorithm. (Publication dates for news stories in Google News are a different story.)

[tangor]

All our pages are regenerated every 6 hours (if updated) every 7 days when they are not, so sitemaps will indicate fresh date. The date on the Guide is usually within the last year. I'm still hopeful.

Might be good for the site and such, but makes no difference to the SE's as they will always compare the "new" with the "cache" and if there's no difference, there.... wait for it .... NO DIFFERENCE!

Might save some server time and regen only pages that actually have changes, since the URI is the same in all cases ... right?

[paranoid android]

I think it *does* have a very significant effect on SERPs. The proof is in the pudding.