|Migrating Site - Will I Lose SEO Positions? |
I'm working on new site design (using WP) and trying to determine best way to keep from losing my SEO positions for certain keywords. I'm ranked high for some of these...
Am I right in assuming that if I create exact same path for my ranked pages that I will not lose these positions?
I'm mainly concerned with my top positions in Google.
There are also many images that are indexed.
Any help would be appreciated!
welcome to WebmasterWorld, Rusty!
are you changing content? urls? internal link structure?
Thanks for reply Phranque - I will be adding new content after move. No plans to change urls... but may optimize internal link structure later.
I've been out of the loop with regards to new changes and techniques for internal linking strategies. This is why I am asking... should I keep existing structure intact? Then optimize later?
I'd prefer to start from scratch, but then I'll lose SEO positions. Don't want that to happen!
As long as your speed stays the same or improves and everything is the same, you have no worries.
Just migrated my 2-YO site to a rocking fast pure SDD. Has about 10,000 pages. I migrated 2-3 weeks ago. Speed improved by 40%, so did my rankings. My google referrals are up 15% since the move with no other new factors, but speed.
|I migrated 2-3 weeks ago... My google referrals are up 15% since the move with no other new factors, but speed. |
No new factors or changes to factors on your side, but the number of updates Google makes to the algo over a 2 to 3 week period of time averages somewhere between 19 and 29, so I think it's important to take many factors into account prior to a move, speed being one of them, but correlation is not always equal to causation, so personally, I'm not "sold" on your increase being based on moving to a faster server, unless your site was snail slow before the move.
[edited by: JD_Toims at 12:57 am (utc) on Oct 31, 2013]
|I'm working on new site design (using WP)... |
Is the existing site also in WP? If you are making a major structural change under the hood there's more of a chance of rankings being affected by that, at least temporarily, than a server migration. I did a few recently with no affect on the sites, except for one! Now I have a bunch more coming up next month and December. I have no concerns. Just expect the unexpected in relation to server variables such as paths, depending on what you may have plugged into WP.
I've reread your OP a few times and not sure if you mean you are migrating to a new server or a new platform on which your site is designed.
In my experience, at the moment, it is not possible to guarantee anything with Google and images even when using the same url path.
Thanks for replies everyone,
Pjman - Speed is good and will stay same
Sevencubed - existing site is older and html based. I guess it's not really a migration - I'm keeping site in place on same server. Just changing platforms.
I will keep same path for each page that is ranking well and hope for best. I guess that's all I can do? Good to hear that status on most of your sites stayed same.
RedBar - good to know... I'm not going to worry about it. I have many images that are indexed, and if they need to be re-indexed, so be it.
I do have another question...
If my site has this path:
Do I have to keep the 'index' in the path?
By changing to WP platform is this possible and advised?
[edited by: goodroi at 4:33 pm (utc) on Nov 1, 2013]
[edit reason] Please no specific urls [/edit]
Okay we're getting places. If your site was previously "old html" but you are now changing to WP you might have an issue with WP dropping file extension names even if you are keeping existing paths. That is a potential problem.
I'm not an advanced WP developer (nor will I ever be) so it's best to wait for feedback from others concerning that. Maybe WP can be configured to allow the filename extensions to remain, I'm not sure.
It's also resolvable with Apache ReWriting but that's a tough nut to crack.
|If your site was previously "old html" but you are now changing to WP you might have an issue with WP dropping file extension names even if you are keeping existing paths. That is a potential problem. |
Personally, I'd seriously consider dropping the extensions while changing management systems so there's no "future issue" with them, especially as more and more sites move to extensionless, which will probably become the "norm" over exposing underlying technology, but it might also be worth waiting for someone who's changed URLs recently to chime in since it's a bit different than simply changing the back-end and how the information is put on the page.
I haven't had any real issue with "small intra-site" moves of that type, but things change constantly and there could be some issues I'm not thinking of right now you would need to make sure were addressed since you'll be running WP and not coding your own solution/.htaccess.
I'm really hoping someone with direct experience with WP platform changes will add their perspective here.
I can't afford to see my site disappear or get sandboxed because I made the wrong move here. My gut feeling says It'll be okay. I'm still a few weeks away from uploading revised site.
Appreciate your input guys.
btw - I'd prefer dropping the extensions - makes sense.
You would like to do three things:
- move the site to a new platform (Wordpress)
- add a new content
- change the internal linking structure (presumably navigation)
And you want to preserve your good ranking (of course).
There are few things to assess and you will have to balance lower risk but less sustainable option with higher risk, but more future proof option.
A) Lowest risk, but least sustainable
The lowest risk but the least sustainable option is to move the current site to Wordpress, keeping the URLs the same, the site structure the same and internal links and navigation the same. This also means keeping the method on how main navigation operates the same e.g. if you have secondary menu open "on click", do not change it to view the secondary menu "on hover" (or vice versa) as this changes internal link structure.
For this option, you would not need 301 redirects (apart from perhaps home page), so there will be no (estimated 15%) link juice lost owing to redirects. Home page will probably need redirect so that it uses domain root and not index.html.
Once the moved site is re-crawled, and hopefully your rankings remain unchanged, wait a few weeks at least to ensure ranking is stable and then you can work on new content and/or internal links structure as a next step. If changes to internal links structure has adverse effect, you can revert in a hope that previous rankings will return.
B) Medium risk, a bit more sustainable
The steps as above, but change URLs to be extensionless (i.e. without .html). Create 301 redirects from old URLs to new URLs. There will be some link juice lost, but from my experience it does not affect rankings much. You will have to make sure that internally you are linking to new extensionless URLs.
Again, you can execute new content and internal navigation changes as next steps once your ranking after the move stabilises.
You will be here in a better position because you have extensionless URLs (on the surface, technology agnostic), but you will put Google through at least two cycles of site changes, once when changing URLs and second time when changing internal linking structure and adding new content.
C) Higher risk, but longer term sustainable
This would be what is normally classified as "site redevelopment". You change at the same time URLs, internal linking structure, navigation and add a new content. It is difficult to predict the results, but if executed technically without any errors, if the structure has improved, if the new content is of a good quality, then *usually* this does not bring long term problems, but it will almost certainly bring short term ranking fluctuation. How "short" the short term is depends on the size of your current and the new site, i.e. how long it is required for Google to digest changes, process redirects, evaluate new and changed pages and so on.
D) The highest risk
The highest risk is site migration where the execution introduced technical errors. For example, Wordpress has the habit of leaking URLs with various parameters that result in duplicate content pages. These will have to be handled by robots or redirects or noindex. Leaking huge number of these at once because of a mistake during a migration can tank the site almost instantaneously.
Whichever route you decide to go, try to make sure of the following:
1) Buy a throw away domain to have as a test site. Block it from Google by either password protecting it or by serving 403 Forbidden to bots (or better, allow only your own IP if you have a static/sticky IP). Least preferable option is to use robots.txt Disallow: / but better than nothing.
2) Decide on the strategy A/B/C above and then install the wordpress on that test domain. Move the content to Wordpress. Change URLs if required or enter permalink URLs to match your current site if not changing URLs. Make sure you are using root-apsolute URLs in your content otherwise your test domain will not truly be test domain. Hence check all content that URLs start with / (and not with http://example.com/etc If you are using ../../ URL format, get rid of it and again, use root apsolute URLs.
3) Use the same images/root absolute image URLs on the test domain as you have on the live site. This way image URLs will not change
4) Once test site is finished, use a site crawl tool to crawl your site (see 2013 Favourite SEO Tools [webmasterworld.com] and Favourite SEO Tools [webmasterworld.com] threads for reviews of useful tools you may want to use for this). Crawl both sites, the current live one and the test domain. Compare the results. If changing URLs, create list of URLs from live site, change the domain name to test site and crawl this list against the test site, ensuring that the redirects are implemented correctly.
When crawling, make sure you crawl with robots.txt honoured as well as with robots.txt ignored, to get the full picture.
5) If there are new URLs/URL patterns to be blocked on Wordpress that you do not have currently blocked on the live site, upload robots.txt on the current site WITH NEW PATTERNS ADDED at least a few days before the actual migration. This is because Google may cache robots.txt for up to 24 hours and putting the new site on the same domain live may result Google seeing pages that it does not know are blocked because it did not re-request robots.txt yet.
6) If you have changed URLs (gone extensionless), create a new sitemap.xml
7) Once you are happy with the site on the test domain, copy the complete content on the live. Pay attention that you have not by mistake copied over test robots.txt and in that way inadvertently blocked the site. Also upload a new sitemap.xml Your .htaccess from test domain will need to be changed slightly to replace domain name from test domain to your domain in .htaccess conditions and rules - be careful of this. Also make sure you remove 403 or password protection from test .htaccess before uploading it to live.
8) CRAWL your site again with a tool as soon as it is live - to make sure the URLs and response codes are as expected. If you have changed URLs, crawl the list of old URLs to make sure they redirect in one step
9) Once you are happy - and if you have changed URLs, go to WMT, perform Fetch as Googebot of the site root and submit the page and all linked pages to index. Perform submitting a new sitemap to Google.
Note: when testing redirects, test both www / non-www URLs from the old domain, to make sure redirect is performed in one step.
Oh great! Now you're suggesting that I become a real webmaster... Guessing I have some homework to do. And I do have more questions... but later.
Thank you Aakkman!
|Oh great! Now you're suggesting that I become a real webmaster... |
LOL -- Happens to the best of us ;)
I have transferred several HTML sites to WP over the past few years, and as long as you follow best practices (and 301 everything important) you might suffer a drop for 3-4 weeks, but it should come back. All things being equal (of course they never are)
Six months in, all my sites were doing way better than ever.
Make sure you block anything in WP that could be considered thin or duplicate content; I block the category pages, the tag pages, author pages (I'm the only author) and a few other things, just cause there's no reason to index them.
I do have another question - Last year I added a WP blog to the site. Path looks like this.
sorry, should have mentioned this... but I was too busy thinking about rebuilding the whole site and this blog has been neglected. Maybe this is an advantage after all? I've been getting zapped by google and seeing drops in ranking. I can understand some of the reasons.
A) Flash movies on quite a few pages - including home page
B) Flash navigation buttons throughout
C) Old backlinks (used linkmaster years ago)
So far, I am in process of removing all flash elements, replacing w/ static. Also removing old backlinks and 'link to us' pages. Almost finished. The site is not too big... so it's no big deal.
I'm trying to avoid further slides in ranking - Now on pg2 for a few KW's where I was in top position for years. I have seen a bounce back to top recently... but seem to be sliding again.
Any suggestions considering this new info?
So if I understood well, the site you want to move to WP consists of static html pages and WP blog in the /blog directory? And you want to move it all to WP?
What is the format of URLs in WP /blog section, are they extensionless or do they have .html or something else?
In general, the steps above do not change, but you have more work to do with your WP install.
I would imagine that when you are converting, you would use WP "Page" for the static (.html) pages of the site, and WP "posts" for the blog part. If your current WP blog URLs are already extensionless (and this is the route you want to go), you would have to make sure that on the re-done site posts have /blog/ in the URL.
You may be able to use your existing WP install to just add "pages" for your static pages, but in that case you would need to move WP install and .htaccess lines that are currently in your /blog/ folder to domain root .htaccess
In fact an average PHP programmer may be able to accomplish the above fairly easily.
With the added complexity of the above, I would stress even more on having test domain where you can complete your site changes. I would copy your entire site (including blog and WP, also including the copy of the WP database) to test domain, change .htaccess and WP install to root to cater for WP and try to create a "page" for the home page and one of your static pages.
Providing this works, you need to see if your current posts in the /blog/ folder works. Note that moving WP .htaccess to root may result in need to tweak the current blog URLs as they do not start in /blog/ folder any more - I am not sure in this though, you will need to check this out.
The alternative is to install the new wordpress in the domain root, use "page" for .html and use /blog/ folder for posts, then copy/paste the content of the current "posts" to the new install on the test domain. In this case I would make sure that my WP database is called something else than the WP database you currently have on live domain, just to avoid unexpected results in messing up the data when copying from test to live.
When happy with the test domain, copy everything to the live as in my previous post, and also do not forget to copy (dump/reload) the WP database from test to live.
With regards to ranking drop:
This should not be the reason, unless the page has very little other content
|Flash movies on quite a few pages - including home page |
I am not familiar with flash navigation in static .html (if this is what you mean?) other than I think SE cannot read them - perhaps someone who knows more on flash navigation can comment.
|Flash navigation buttons throughout |
It would depend on how many they are within your complete link profile
|Old backlinks (used linkmaster years ago) |
But regardless how you approach it - do it on test domain first, and then create a list of steps to execute when moving a test site you are happy with to live.
Considering a new info on blog, I would personally probably choose the route on rebuilding the site
Thanks for info AA - it makes sense working on test site.
path looks like this in blog /blog/page1/
No html extensions
Yes - Flash navigation on static html pages.
|Flash navigation on static html pages. |
"Flash navigation" jumps out at me as a recipe for disaster. Please clarify.
You really want to use search-engine-friendly html navigation, preferably text.
Site has Flash nav -
I'm removing it now... almost finished.
Could this be why google despises me?
Could be a large reason, depending on how link juice is flowing through your site, but there may well be other reasons too.
I'd let Google absorb the changed nav before making other changes. I like to isolate factors as much as possible, and this is one where it's easy enough to wait a while and see what the effects are.
Your mention of index.html suggests that canonicalization is likely to be necessary.
Will do. I have time... In fact it's better this way. Still a few weeks away from bigger changes.
Thanks for tips. Reading up on canonicalization and 301 redirects - new terms for me.
btw - there have been text links for navigation in footer.
If your navigation has been duplicated as text navigation in footer, then your link juice flow should have been fine.
But I am confused. Reading your opening post:
|I'm mainly concerned with my top positions in Google. |
And the very recent post:
|Could this be why google despises me? |
Then I am not sure if your site ranks well and these are rankings you want to preserve, or ranks badly?
The evaluation of risks is different in these cases. If it ranks well, you are aiming to preserve good positions, hence various options on how to minimise risks.
But if Google does not like your site, you have not much to lose - in which case I would go for the complete site revamp during your move to Wordpress.
The search rankings are wavering between pg1-2, a big difference for my business... I was exaggerating a bit... Google has been playing with my site, but not enough to do a full revamp.
It's a fine line - If I make big changes, then I have to be prepared for no income.
So I have to be careful.