Using Noarchive to remove your cache – what’s the impact?

by on 13th November 2009

Recently, I’ve been interested in understanding if implementing an instruction to remove the cache link using the “noarchive” tag has any impact on your search engine traffic and rankings.

As Dave puts it:

So the only reason I can see why I should leave the Google Cached link is so that other people can see what my website looks like in Googles eyes and why would I do that ?

I tend to agree. In fact, why is that cached page link there? It hardly inspires confidence to most (ordinary) visitors if your site is down, so the common reasoning from Google behind having a cached link available seems weak at best.

What are the alternatives to the cached view if you’re an SEO?

It’s easy enough to come up with reasons why the Google cached view is useful to an SEO, but there are other ways to replicate a similar, if not exactly the same view of a page if you’re doing on-page diagnostic work.

web developer toolbar

Viewing a page in Web Developer toolbar with CSS and Javascript disabled has been my favourite way of viewing webpages as a search engine might for a long time. You can also use the page analysis button on the SEOmoz toolbar, though I much prefer a combination of Firebug for viewing code with Web Developer controlling the view.

Other options include viewing your site using Lynx or SEO Browser, though I’m all about doing SEO with Firefox and Chrome.

“You’ve missed Fetch as Googlebot tool!”, I hear you say! Apparently, the tool has limitations. If you think about the technology used to fetch a HTML page and render it from a cached copy in a real crawler “user search” situation, it’s hardly likely that Google’s fetch tool would exactly replicate their real life crawl. Not to mention that, if the tool was a perfect replication of a Google crawl, it might expose security problems or flaws in Googlebot’s base code. Cool.

Now you see it…

SEOgadget in the search results pages with a cached link

Now you don’t…

SEOgadget without cache link

For purposes of pure interest, I added the noarchive tag at 3.30pm, Wednesday 11th November. I first noticed the cached link had gone at 9.43am on Thursday 12th November. Of course, the cached page delivered via the “cache:” query in Google still appears (4.12pm Thursday):

seogadget cache link

It’s interesting to point out that, regardless of the noarchive instruction, the cached URL of my site still works with a current cached date of 13th November at 08:21:19 GMT.

The impact on SERPS and traffic

There has been no impact on my search results and traffic remains quite healthy. I’ve been monitoring my top 100 keywords in Google UK, Bing and Yahoo for the last few days. Needless to say, nothing has happened yet and I genuinely doubt that it will. Of course, if I discover anything then I’ll be sure to let you know on Twitter.

As with all things SEO, I advise you to do your own tests. Just rest assured that implementing the noarchive tag in your header will not hurt you in the short term, at least.


  1. Several papers use it with no ill effects. The Mail and Telegraph (news stories but not blogs) both have noarchive on their pages.

    This also lets them change stories with no obvious sign they’ve done it (EG think the headline during the Jan Moir affair).

  2. Nice study Richard. I agree with you and Dave Naylor – don’t really see why I’d want to invite others to see my cached page. Good to know that the cache: command still works just the same – wasn’t sure about that one!

  3. Good test Richard, and I’m interested to see if Google eventaully drops the cache: access.

    If I remember rightly, and I could be wrong as my memory is getting bad the older I get, I seem to recal that the Cached copy of the page will eventually stop showing for the cache: command but that it takes about 7 days longer for this to happen. Watch this space as they say.

    Like I say, I “think” thats what happens. I checked something like that a while back (3 years ago) when I was trying to game access into expert exchange answer pages.

  4. On the cache command, here’s a page from the Mail that has always had noarchive on it and is a couple of weeks old – you can see it via cache command. On the other hand, there are old pages I can’t get to appear via cache. Not sure what that means!

  5. Malcom

    If what you say is correct, then either it takes longer than 7 days (the figure I mentioned) or less likelly, Google has changed the way it deals with the cache: command recently and not applied that change retrospectivelly.

    As Richard is running this test we will soon have an answere.

    Good. More like this please.

  6. Several of the newspaper and magazine sites that I’ve worked have implemented the noarchive tag on all of their content. To-date I haven’t seen any negative (or positive) effects from its use.

  7. Fascinated by the fact you can override the site’s wishes and use the cache command anyway. Sorry for all the links, but I just wrote this about two newspapers pulling stories, probably because of a press complaints commission complaint.

    However, you can see the two original stories here and here using the cache command – even though the Mail uses noarchive.

  8. I’d make some tests and I actually do agree that it won’t affect any rankings.

  9. I have seen many ways and tools to help see a page as search engines would see it. Wouldn’t it be easier to just view the Google catched page: and then click on Text only version. This is my favourite because I can easily see all the text the Google sees including alt text etc. Of course won’t work if cache is disabled but I don’t’ come across many sites that have it disabled.

Comments are closed.

Get insights straight to your inbox

Stay one step ahead of the competition with our monthly Inner Circle email full of resources, industry developments and opinions from around the web.