On Page vs Off Page - Which Has a Greater Effect on Rankings?

[Travis-W]

Hi Mozzers,

My site will be migrating to a new domain soon, and I am not sure how to spend my time.

Should I be optimizing our content for keywords, improving internal linking, and writing new content - or should I be doing link building for our current domain (or the new one)?

Is there a certain ratio that determines rankings which can help me prioritize these to-dos?, such as 70:30 in favor of link-building?

Thanks for any help you can offer!

[SEO5Team]

Hi Zora,

My thoughts are that it really depends on the competition in your niche. If it is not very competitive then just the on site optimization itself should bring you good results. (quality , optimized content being key). If ,on the other hand, there is a lot of competition for the keywords that you are trying to rank for , you will have to do the whole bit (on site and off site) to get traction.

[William.Lau]

Off-page optimization is more difficult hence it is much more powerful in rankings.

On-page SEO should always be the first thing you do because you can control it and easily fix and adjust everything. Off-page is more difficult since it requires partnerships and PR, which of course you can't control that easily.

[ThompsonPaul]

Here's how I look at it, Zora:

• On-page optimisation is now simply the price of admission. If you want to play effectively in search, good on-page is the foundation of everything else you're going to do. If you've got inadequate on-page optimisation, you're going to automatically compromise the effectiveness of all the off-page work you're doing.
  • For example, good links to a page with confusing meta-data and poorly organised content are not going to bring much value. Plus they'll be a LOT harder to get in the first place.
• Really strong, high-quality, useful content will make it MUCH easier to attract links, whether naturally or through link building.
• 301 redirecting is considered to never pass quite as much "link juice" as a native link pointing directly to the landing page

With these things taken into account, my workflow in your situation would be

1. to ensure the on-page optimisation for the new pages on new site is as strong as you can make it
2. take time to research and write new, strong, linkworthy content for the new site
3. do some solid advance planning and have an outreach plan ready to start trying to attract links to the new content as soon as the new site is up.
4. any time left over, pre-plan further off-site link attraction strategies for the new site.

All this assumes you have an absolutely ROCK-SOLID plan for redirecting the URLs from the existing site to the new site, and a foolproof, tested implementation plan for those redirects (plus a solid strategy for monitoring and adjusting them instantly should any issues be discovered).

Does that make sense?

Paul

[WhoWuddaThunk]

I agree with William.  Offsite is the single most powerful aspect of your SEO, but it is just one part, and shouldn't be your sole focus.

I would prioritize like this initially:

1. Optimize current pages

2. Fix errors (404's, etc)

Then cycle through these:

3. Write New Content

4. Internal Link Building

5. External Link Building

So, you do the initial set up of your website, then go through the process of creating content, and building links.  How you do the second part is really up to you.

• You could write the content first, and then try find links.

or

• You could find content that needs to be wrote, write it, then contact those people with insufficient information to link to you.

Either way the content is the medium you need in order to be most efficient at link building.

[Travis-W]

Everything makes sense up until you got to the ROCK-SOLID plan.

Shouldn't I just redirect all the pages to their matching page on the new domain?
For example siteA.com/article3 --> siteB.com/article3

We may be removing pages, in which case I will either redirect them to the closest relevant page or the home page.  Sound like a solid plan?

Also, one question about 301 redirecting.

Say I redirect  siteA.com/article3 --> siteB.com/article3 with a 301 redirect,
and I realize I made a mistake and need to redirect it again to another page 
siteB.com/article3 --> siteB.com/article4

Do I lose another ~10% of link juice, or is it just considered a single redirect from
siteA.com/article3 -->  siteB.com/article4 ?

This question does not necessarily apply to just between domains, but 301 redirecting in general.

[ThompsonPaul]

Let's deal with 2nd question first re changing 301s. As I mentioned in your more recent question, the solution is to go back to the original redirect and rewrite it to go to the new correct location.

As long as you still control the original redirects, (which is why you'll want to maintain your old domain for at least 180 days, preferably permanently) there's almost never a reason to chain redirects together like that. Always better to rewrite the original to the new location. In addition to a number of technical benefits, it makes those redirects VASTLY easier to maintain.

As for a solid redirect plan - yes at its most basic it's a matter of ensuring you've catalogued ALL the old pages and written good redirects to the new pages. But there's plenty more to do if you want the transition to go really smoothly!

For example - as in the info above, you'll also want to check the old site for existing redirects and make sure they get included and rewritten as one-hop redirects to the new page.

Perfect example of this: original site had several vanity short URLS redirecting to landing pages. E.g. www.site.com/sale was redirected to www.site.com/category/colour/somereallylongurl

You not only want to redirect the /somreallylongurl page to it's new equivalent, you need to make sure you've also changed the /sale redirect to point directly to the new page as well (not having to go through /somereallylonurl to get there). Really look for these - there are often more of them than you think.

You'll want a quality assurance plan in place to test as many of the redirects as possible just as soon as the new site is live. One misplaced or forgotten "?" in a RegEx can completely foul things up. This is a daunting task, but it's essential to catch as many mistakes as possible before the search crawlers index the goofs (or your users run into them!). You'll probably want to divide this task amongst a group of testers. Watch especially for whether redirects resolve whether the address of a page ends with a "/" or not.

You'll also want a solid plan for how to ensure the search engines find and index those redirects and the new pages quickly. This is sometimes done by placing the old sitemap on the new site for at least a few days so that the search engine spiders are forced through the redirects for all the pages. Not everyone agrees with this approach, so you may want to research it further.

Here are several additional things you will absolutely need to do in Webmaster Tools.

• As soon as the new site is live, set up and verify a new Webmaster Tools account for Bing and Google. (You can even do this in advance using DNS verification method or if there's a placeholder page live at the new URL before the move)
• As soon as the new site is live, use the Change of Address tool in Google Webmaster Tools to point to the new domain.
• Use the  "Fetch as Google" in Google Webmaster Tools to find and submit a series of pages in different sections of the new site. This step helps the engines find and start indexing your new pages faster. Note that in Google, you'll want to request "Submit URL and linked pages". You can do 10 of these a month, so pick pages from different sections of your site for best effect.
• Use the  "Fetch as Bingbot" in Bing Webmaster Tools to to do the same thing you did in GWT above. Bing allows you to do 10 of these per day. and a total of 50/month. Use all of them, again from different sections of your site.

You'll also want to make certain you have an effective robots.txt file in place to ensure the search engine crawlers aren't wasting their crawl budget on useless pages.

Here's something else to think about. Can you get the owners of the sites that are giving you your most valuable incoming links to update their links to point directly to the new equivalent URLs? Doing this for even 10% of your top links has multiple benefits. The engines will find your new pages faster via the authoritative links, and those links won't lose any "juice" by having to pass through a 301.

As mentioned in my first reply, now would also be a really good time to get a few new links to the new site from some strong external pages.

You also need a solid plan for what tools you will use to monitor for missing/broken redirects and how you'll fix them quickly.

• Webmaster tools to watch for new 404s from incoming links
• Google Analytics (Make damn sure you have your tracking code included on your 404 page - many sites miss this so don't even see all the 404 hits).
• Open Site Explorer as a backup to Webmaster Tools for spotting broken incoming links
• Screaming Frog for an internal link scan to make sure you didn't create or inherit any broken links yourself accidentally
• When will you scan?
• Who will rewrite the .htaccess if fixes needed? (You don't want multiple people in there messing about).
• Who's responsible for updating the sitemap.xml file to the new one after a couple of days and resubmitting it to Google and Bing Webmaster Tools?
• Who will check GWT's Sitemap Details page to ensure the majority of the site is getting crawled and indexed, and will know what to do if it's not?
• Who will follow up with a search of the engines in several months to see if any of your old site URLs are still indexed and insure redirects are in placer for them as well?

The whole point to all these steps is to use every method at your disposal to insure both Google and Bing fully understand the relationship between the old & new site, and to push them to transfer the old site's authority, rankings and traffic to the new site as quickly and efficiently as possible. And even though Bing may send less traffic, you still can't afford to throw it away, so you'll need to monitor it's own Webmaster Tools as well as Google's.

Wow - I know that's a helluva load to think about so be sure to ask if there's stuff I haven't made clear enough.

Paul

P.S. The last thing is to accept that things will likely be a wild ride in the SERPs and traffic for at least two to three weeks after the move. Don't be tempted to overreact and start making large changes too soon if things look wonky at first. The only exception is 404s - fix those as soon as you detect them.