Forum Moderators: Robert Charlton & goodroi
Cheers
Another site did just fine for me too. This one was a complete overhaul and the site changed all of it's conent too. I kept the original site and URLs up - so that if someone still visited site.com/thispage.html it was still there and visible. However, this was a part of the old site and not to be a part of the new site. On the front end the site was 100% different and it too kept it's listings. So when you went to site.com - everything was different. (Although the old site is still up and running - just not linked to from the main page.)
I'm always nervious when doing such a thing - but my experience has been if you plan your redesign correctly, that you won't have a problem. I've also moved sites between locations without a problem either. This too scares me, but in the 5 site moves (from one IP address in Virginia to another IP address in San Jose say) I've not seen a problem either.
We spent thousands on a new design, launched it on Jan29th/Feb 1st. With the Google update on the 2nd, we were slapped with some kind of plague that prevented us from ranking well... Ironically, sales are normal, so obviously Google traffic isn't a huge factor (even though our traffic dropped over 60% once we lauched our redesign / got dropped in Google).
If you are reliant on you Google traffic in any way, I'd suggest holding onto your old design. Unless of course it's just atrocious and not converting at all - in which case the redesign would be recommended.
We didn't change URL's or architecture, just design.
I should mention - the redesign was completely for a users perspective. We did enough SEO such that it should have been at least on-par with the old design, but no SEO magic or scams. We have too much good content to bother anyway, it would be counter-productive.
My theory is that if you make dramatic changes, they see the site as having changed hands or something and put your rank back to 0.
Page names were not changed.
Don't do it unless you can justify the loss of g traffic.
However, I have seen sites that had certain "coding errors" benefit from "clean ups". But that is a different story.
Yup, I'm hearing lots of rumblings of re-designs getting smacked.
Have you heard anything as to the TIME to getting smacked from re-design of website?
I persoanlly thought 2 days was a bit fast, but we're a pretty big company, PR6/7, thousands of pages of content, so we get spidered a lot. I guess 2 days was enough!
I don't see what difference it would make to do nothing but change to nicer looking graphics and/or colors, but even then there could be a change of image names, which can possibly make a difference.
>>Don't do it unless you can justify the loss of g traffic.
I've got one that got smacked by Google, and that one is worth changing because it doesn't get any Google traffic any more anyway, and it would be a great one to test to what kind of changes would help. But otherwise, things seem too unstable to do anything but correct errors, fix any broken links and clean up code a little.
... too unstable to do anything but correct errors, fix any broken links and clean up code a little.
I must remember this quote next time someone says that Google isn't hindering development of the web!
Redesigns are usually to benefit visitors, so I find it strange that they smack sites that update - however it appears to be true.
Do they re-enter the sandbox, or get eliminated for less time that than.
Also, again - how long from time of changes to time of being smacked?
Before the latest dance, we had thousands of keyword combos that produced top SERPs. At this point most of those phrases dropped 10 to 50 spots, many are gone altogether. We went from 12,000 visits per day to about 700.
In my situation the damage occured within 48 hours after the redesign. Unfortunately, because this coincided with the latest dance, it is hard to know if the drop is due to the changes we made, the dance, or both.
For two months no problem, nothing changed.
After two months I had a "strange" experience; I lost my rankings for my preferred targeted keywords (from page 1 to page 2 or three) but I gained traffic. I gained traffic for several keywords that are not so useful for us.
We had to increase a lot our link popularity to regain position.
With another site we moved to CSS and the transaction was fine: we gained traffic and ranking!
My suggestion is:
- Try to use CSS
- Increase your link popularity before, during and after the moving
Link popularity is like a vaccine....
The site is very dependant on Google (approx 90%) and has grown to a substancial level of traffic over the years. Thus my concern.
I think it will start with small css changes. Test the waters so to speak.
prairie, are you referring to a redesign with URLs changing, or remaining the same?
In my experience over 3 old sites, changing URLs leads to some deep freeze time, but I think the real issue is the extent of change.
I've changed content on existing URLs without issue (on old sites).
Others have reported effective ranking demotions courtesy of adding a lot of new content at once.
I think the lesson is to plan your site's structure/file names well ahead, and build in room to grow. Unless you have a large site, or maybe a very solid domain, add content cautiously.
Its a pain, and not really fair, but that's Google these days (for now).
The one (perverse) exception I can think of is with a small travel site which Google dislikes - despite a lot of original content and images. For some reason best known to itself Google delivered up to a hundred visitors a day to a page that can only be described as complete rubbish. It was nothing more than a holding index with a few internal links. I felt that it reflected badly on the rest of the site so I added some relevant content. In the recent update, Google has now dumped that page into the depths. C'est la vie.
Search traffic is now zero after the recent update. Can't find the site for any search terms, however if we type in the company name we show #1.
My client though does not depend on SEO as the brand is regionally well known and counts for about 90 percent of the traffic. Still I felt it note worthy to share traffic results from Google are zero.
Daily Googlebot visits and does a complete crawl.
If you're going to rename links, use 301 redirects. We did not do that instantly and I regret that. Our site is online for five years now with good clean content (mainly how tos and tips) and we changed URLs several times. But this time we really have been hit hard.
So be careful -- think about the good old saying "never touch a running system".
All we can do now is wait. Googlebot is spidering madly and we hope to be back next month.
[edited by: lawman at 9:41 pm (utc) on Feb. 12, 2005]
[edit reason] No Url Drops Please [/edit]
As soon as the redesign was finished (and sometimes before) they all took off. I rarely changed file names but often changed some of the content and especially made sure keywords were in the title, meta tags, headers, body text and bolded text and placed an intro paragraph full of keyword rich text at top of page. Two sites started getting orders before the site was even finished.
If your site didn't have those problems maybe it's due to too much competition for your keywords and you need to payperClick.
. I rarely changed file names but often changed some of the content and especially made sure keywords were in the title, meta tags, headers, body text and bolded text and placed an intro paragraph full of keyword rich text at top of page.
did you change the layout of elements on the page?
Eg. move the navigation menu. I am keeping all content exactly the same, just improving the visual aspects.
Let's assume that someone does a redesign of their established ,well ranking site... but only forwards traffic there by detecting browser, and sending people there who are IE 5.5 or Firefox 1 or higher.
85% of your visitors would see the new site...
Googlebot, and other spiders, and people with old nasty browsers would see old site (which would continue to be viewed as unchanged by Google).
Would this be considered "cloaking" or is this acceptable since you're forwarding new browsers to the new layout, and old browsers to the old layout?
--Mark