The classic advice came from Tim Berners-Lee in 1998.
"Cool uris don't change": [w3.org...]
|Basically they said prove to me why I need to |
Ouch! It sounds like there may be a problem with your relationship with this client. If they pay you for advice, the cornerstone of that relationship is them trusting the advice you give. Otherwise, everything could be hard work. It's impossible to "prove" most SEO techniques - you can provide evidence that it's worked on other sites, but rarely anything much more than that. Are they saying they want someone else's opinion before they'll listen to you?
Imagine a different scenario - if you were a mechanic, say, and advised someone they needed to change a particular component. If they said prove it, you'd likely let them walk away ;)
Certainly, it makes no sense to leave redirects in place on internal navigation. I'd probably let old articles and the like slide (depending on the number of them) but anything header/footer/repetitive I would change. The major reason for me is that redirects don't necessarily pass ranking signals (juice), whereas the chance is considerably higher with a straight href link. So leaving redirects on links you can control is risky, for something that's typically straightforward to fix.
You need inbound links.
Lots of good ones.
There are two types of inbound links.
Internal links and External.
If you changed the page url, the old page is no longer available.
Now, without changing the internal link structure, you try your luck to perform well without internal links.
It might still work, but it's not wise.
Thanks for the feedback. Actually they have been a very good client and take advice well. It is just a few of the programmers/designers that are not buying in where as management has. Either way I still would like to present some evidence if possible to them.
The bottom line is it is very hard for them to technically make this change. Time consuming and difficult. I made them aware that they would have to from the beginning, they were just looking for proof that this change is necessary right now. Which I feel it is as the new pages are live and now most of the navigation is 301 redirecting to the new pages.
God, just tell them that they are idiots and that's why they are paying you ;)
Seriously, I know of a case just like yours (major hotel chain that did this for its national website in one country) and dropped significantly until the redirects were sorted out. Tell them you double-checked with other webmasters, including serious ones such as wolfadeus from ww :)
It seems to me that you are trying to convince the client that best-practice is critical when, in fact, it probably isn't.
If there is no easy way to update the links and the redirects are working correctly, then, from the client's point of view, the money can be spent better elsewhere.
From a search-engine perspective, there should not be a problem - they index pages not websites. They don't much care whether a link is internal or external. Redirects mean more work for the spiders but no penalties seem to accrue.
> They don't much care whether a link is internal or external. Redirects mean more work for the spiders but no penalties seem to accrue.
@kaled - do you have proof to this assumption?
> I know of a case just like yours... dropped significantly until the redirects were sorted out.
Me too. I know (but sorry, will not be able to provide proof) that internal links are part of the basics in a linking structure, sometimes (as in the examples I know of) an unused gold-mine by lazy webmasters/programmers.
> the money can be spent better elsewhere.
It can be spent elsewhere.
|They have 301 redirected the old URLs to the new ones properly, but are hesitant in investing the time in changing the internal links (navigation, etc.) to match these new URLs. instead they feel they can just leave the redirects in place and everything will work out. |
It probably will. I wouldn't leave it to chance though. That type of attitude might be fine for some low end stragglers but not for primary navigational links.
Tell them that link equity is fragmented in the redirect. That should open their eyes. ;)
From a tech point of view it might be difficult to explain why internal linking should be fixed. You know it should, I know it should, but management might be skeptical.
Try this: "301 is just like telling a lie. The page the user asked for isn't there so we lie and tell them it is over here." After awhile the liar will eventually screw up and forget what lies they've told and then...
Link Checker - 301 > 302 > 302 > 200
We're discussing how redirect chains can and will cause challenges.
Q: Do I have proof of my point?
1) GoogleGuy/Matt Cutts has said on several occasions that absolute urls are preferred.
2) Whilst sites can accrue penalties and might accrue trusted status, it is pages that are ranked, not sites (or so we are told).
3) If the old urls are vanishing from search engine indexes and being replaced by the new urls, then it would appear that everything is working correctly. If this is not happening, you could use this as evidence to support the case for updating internal links.
As I understand it, 301 Moved Permanently is intended specifically for the job you have described.
Kaled, your missing the point. The 301 redirects are in place, but they dont want to change the navigational links, which is a completely different subject.
If you re-read my first post you will see that I am not missing the point at all.
The point that I was making was that if you have set up 301 redirects and they are working correctly, there is little point editing all the internal links because the 301 redirect is intended specifically for problems like yours.
I agree that best practice dictates that all internal links should be direct, but companies have to balance costs against benefits. In this instance, best practice may be good for your balance sheet but not for the client's balance sheet.
"I agree that best practice dictates that all internal links should be direct, but companies have to balance costs against benefits. In this instance, best practice may be good for your balance sheet but not for the client's balance sheet."
Then that is where I disagree as I have seen websites do poorly when managing their internal links in this manner.
|the 301 redirect is intended specifically for problems like yours |
I don't agree that a redirect is intended as a replacement for updating links when the destination has changed. A redirect is for a URL - not a link. Even the spec suggests that links should be updated:
|Clients with link editing capabilities ought to automatically re-link references to the Request-URI to one or more of the new references returned by the server, where possible |
Explicit in the specification of a 301:
"...any future references to this resource SHOULD use one of the returned URIs"
I.e. update the links wherever possible.
From an SEO point of view, the point has been made a few times above - 301s may retain the ranking signals, but they have less chance of doing so than a straight link. This is increased risk for the client's website performance.
Finally, how difficult can it be to update site-wide navigational links? This should a quick and straightforward task IMO, which should make any argument against updating moot.
Sorry to be dense, but if you have compiled a list of OLD->NEW pages (which you must have done for 301 redirect), why can't you just find and replace the old URLs for new?
It is not that simple for some designers as it is not just changing the URLs in the navigation but site wide.
Sure, but the links must be somewhere- either hard coded in html or in a db table.
Is it that you have been using relative URLs? If so, well, that WILL be a problem. I'd tell your programmers to use absolutes (at least from the domain) in future
I have usually found that in cases when I think something is ought to be simple to do and the dev guys tell me it's really difficult to do, the core of the problem is that the dev guys had cut corners to begin with.
This means they have to go back and do it properly, which they don't always like.
[edited by: Shimrit at 3:38 pm (utc) on Feb. 18, 2009]