What to do when you are rebuilding a site

Hello all,

I am redesigning and rebuilding a site after learning some great tips from the DP forums. The page has a 0 PR and had one backlink. I am just wondering what the best way is to go about not having the website become trashed as far as future spidering goes. I have created an under construction page. If it takes me 1 month to build the new site, will G, MSN and Y come back and visit if I leave that page the way it is until I am done?

 

backlinks -- get ALOT of them - Spend an hour a day for the next month trading links.

Also, Get into DMOZ and other directories. Apply for those RIGHT NOW (they take awhile to get into)

Also, resubmit your site to google when you finish the redesign. It never hurts to do so.

 

I don't think he's going to find many people who want to link to an "under construction" page, joeychgo...

NewComputer: Try making that under construction page more of a "placeholder" oage that describes what the site is for, what it's about -- i.e., a few paragraphs with important search phrases/keyword descriptions of what the site will be. Then if it takes a month to finish the site, modify the content of that page from time to time to keep the spiders happy. As you complete parts of the site, add temporary text links to those pages from the "placeholder page".

 

He already has a page - ther new page is being built to replace the old. Hence, he already has a placeholder page.

 

Backlinks .... Yes, that's a very good idea, but only after you have finished the site as not a lot of people will trade links with you if you have a PR0 and an under construction page.

Same for DMOZ listings and other directories. Just in case your site is reviewed before you finish building it, you don't want to be rejected b'cos your site's under construction.

And finally, it is not recommended to resubmit your site to Google as some people believe that it resets your submission date and might put you at the end of the queue each time you do this. Hope this helps!

 

Usually when someone talks about an "under construction" page, that page shows up as the default page, so the previous site becomes "unavailable" or "inacessible" except perhaps by direct URL, assuming the old pages still exist.

So there are two options:

1. if the old site has been deleted, then follow my suggestions above in constructing a "placeholder" page.

2. if the old site still exists, then leave it there, intact, as a whole, until the new redesigned version is ready to upload. Until then, build it as a local disk-based site so you don't over-write the previous pages until you are ready to...

 

Agreed, on both counts.

No. That's DMOZ, not Google.

 

That's true! DMOZ, it is!

 

Thanks Minstrel,

That was just the idea I was looking for. To the others, I may have not made myself clear enough. I just don't want to have a stale 'under construction' page that after a month of the spiders hitting it they decide not to come back. Then I get the new site up and it doesn't get indexed because of the under construction page being up so long.

 

Change the content of the page daily - even if it is only adding a a date and time. Then the Bot sees it is changing and comes back regularly to respider.

SO .... when you new redesign is ready, the bot will be coming regularly and will index the new site quickly.

I made the mistake some time ago of not doing this. I set up a single page to get the site into the index, then left it untouched. Google thought it wasn't updated frequently, so Googlebot came by infrequently, and, once the full site was ready, it took a few weeks more than it should to get the new pages fully indexed.

 

What you seems absolutely right to me. Except that I would like to do it differently. Tell me what you think of my approach:

I am preparing a scholarly site. There is no real reader value in changing content regularly, especially the date / tiem variety tinkering. Hence, I am hesitant to force daily change solely for SEs. So, I think I will live with infrequent spidering. But only as long as the "unupdated" content does not lead to lower ranking (as opposed to simply delayed spidering).

Seondly, I feel that a new set of back links might trigger a quicker re-spidering when I need it. Do you agree with this?

Thanks
Ajeet

 

"Changing content" doesn't nevessarily mean throwing out one article and writing a completely new one. It can just mean making some minor changes to the wording or layout of the page, so that the "last-modified" header on the server for that file changes.

 

Minstrel, is there an easy way to see if the last-modified header has changed for the page? I mean from an internet browser or a tool out there?

 

Yes. Here's one: http://www.delorie.com/web/headers.html.

 

Minstrel,

It didnt tell me if they had been changed or updated and when. That is what I was hoping for. Here is the results I did receive: HTTP/1.1 200 OK
Content-type: text/html

Edit: Weird, some of my sites give the time etc... others don't. Would this affect the SE spiders, postings, indexing? The sites that are not showing time and date are the ones that seem to not be showing in MSN and Y!

 

The output should look like this:

If the output for your page doesn't look like that, I'm not sure. It may mean that your server is not returning the required information.

As to what effect this will have in spiders, I'm not 100% sure. I can imagine that it might affect frequency of spidering the site but beyond that I don't know.

 

Thanks Minstrel,

I do get that on most of my indexed pages, but on the other site is a different story. I will continue to dig on the web. If I find a result I will let everyone know.

 

Thanks, NC -- I'll look too when I get time. Interesting (and important!) question: What does the spider do if it tries to get "last-modified" data and nothing is returned? I would imagine that it will index the site anyway if it's a new one but if that site is already indexed, then what?

 

Yea, I would think 'adios'. If you think about it, the spider would hit your site and look for the norm: Robots etc... and also the last-modified. If it had not been changed what would be the point of it spidering it again. This would save time in getting their database reloaded with more 'relevant' and fresh information. Hmmmm, unless you were using a dynamic page or a server that does not return last-modified headers. Just a thought. I will keep digging. One of the foremost 'Google' masters has said elsewhere that last-modified headers are very important.

 

But that's the crux: If there is no "last-modified" data returned, does it assume "no change" or does it assume "possibly changed"? "No data" isn't the same as "Not modified"...