Welcome to the Q&A Forum

Hey Ahalliday,

I have seen huge structured-data driven sites with 100k+ pages set online at once to no detrimental effect, but I wouldn't do it. Generally speaking you should be save to start with half and see what happens.

What I would suggest is that you split your pages by categories / types or whatever seems appropriate and list each type in its own distinct sitemap.xml . When you turn in the sitemaps in the webmaster tools you can see the indexation by those categories / types. Any low indexation % might be a hint that the pages are indeed spam and not worth indexing.

Sebastian

Hey all,

I discovered that the link from the public profile might not count. To my knowledge Google "counts" the first link to another page on any given page. The public profile has a nofollow link before your follow link:

linktodomain.tld

linktodomain.tld

Doesn't this invalidates all the public profile links? What is your stance on this?

Regards

Sebastian

One of my client tags image-links with nofollow if those links are before a regular HTML link in the source code, e.g.:

.
.
.
Check our page aboutreally cool shoes.

They do this to "better pass anchor tags" to the page /page-about-shoes.html.

My question: Is this a good practice to do?

Thanks

Sebes

I have a client using the following construct throuout the site:

linktext
linktext2
linktext3

This is obviously wrong, but does it work? Does Google / bing recognize all links in the paragraph as nofollow? Any ideas / opinions?

Hey Erica,

thanks for your answer. What I need is a way to decide on-the-fly whether two pages are similar or not. If they are too similar I need to depublish or at least rel canonical one of those.

Best solution would be an API that takes 2 pages, but it seems as if I have to build it myself then.

Thanks for your efforts.

Does anyone know a service or API or php lib to compare two (or more) pages and to return their similiarity (Level-3-Shingles).

API would be greatly prefered.

There is a new German based service called strucr.com which does exactly this: analyzing large sites and their internal linkstructure. But it is mostly for realy large sites.

Other options you have is to use xenu link sleuth and crawl your site by yourself.

Short answer: DONT!

Repair the glitch even if it is a little bit longer. Let me phrase it this way: "redirects are serious business" and not for html problems.

More on redirects and statuscodes: http://www.seomoz.org/learn-seo/http-status-codes and http://www.seomoz.org/learn-seo/redirection

Hey MarieAnne,

ok, so you got 5 different designs for your page and more domainnames that you want to point to one of your designs. This is perfectly ok, you should only watch a few things.

I made a quick check on canadafur.com. It does not redirect with a 301 statuscode, but with a 302 (temp redirect), you should change that to 301 here is more on statuscodes: http://www.seomoz.org/learn-seo/http-status-codes

I also noticed, that furcouncil.com does not have a rel canonical. This is optional, but in a scenario you are describing it realy gives you security not make a mistak. What you should do is: rel="canonical" href="http: yourdomainhere.tld/page" /> in the header of each of your pages. This will tell the searchengines that yourdomainhere.tld/page is the version you want to display and what they should index and rank. 
Why to do this? If you screw up one of the redirects it could happen that the exactly same page is one time crawled and indexed on yourdomain and one time on anotherdomain, thus creating duplicate content.

More on that topic:

http://www.seomoz.org/learn-seo/canonicalization and  http://www.seomoz.org/learn-seo/duplicate-content

If this helped and answered your question I would be very thankful if you flag this answer as helpful If you have more questions just ask or PM

Thanks

Sebastian

I soo am wainting for the first "PM me & we talk" posts

But on toppic: If you need to buy links, than see what domains/pages are also ranking for your targeted keywords and directly contact those webmasters and see what you can work out. Skip Teliad et al. There are bloggers here and there who will link to you for a free sample of your product or some paid article without disclosing it. But beware: you are in violation of Google's webmaster rules.

I agree. I would argue that tld distribution is partly determined by link sources (i.e. what country the page is located in) and the targeted pages language/country. So a strong deviation from that will probably red flag a domain. Exampel: a page hosted in Germany, about Africa recieving tons of links from china might be fishy. A page in Poland about Germany getting links from Poland (or Germany) might be totally relevant, but with differently tld distribution.

EDIT: So the best rule of thumb is to ask yourself "If a human QA rep would look at the page and it's links, what would he think?"

According to Matt Cutts page load speed is not as important as a ranking factor as one may think.

http://www.youtube.com/watch?v=B3zmP0W26M0  old video, there is a newer one from 2011 which i cannot find

Regarding the usefulness of page load speed: I heard about tests run by amazon and google that more or less lead to the conclusion that 1% faster meant 1% more revenue (to be taken with a grain of salt). But it is obvious that really fast pages produce more pageviews (and ultimately more ad-revenue etc.) and also increasing your sites user metrics (time on site, avg PI ...).

Some tips: go over every element and ask yourself whether you really need that element.

As Robert stated I would aim for a average page on an average domain to load between 1,5s and 2s.

Good ressources are:

www.webpagetest.org

Googles own Tool "PageSpeed"

Hey Keri,

I do not have hard data to prove this claim, it's purely my opinion and I should have marked it as such. I had a few presentations to customers recently and most of the time when I asked them to go to a specific page the person would use a hyphenated (spelling?) version or search brand1 keyword1 instead of brand1keyword1.
I personally first try a hyphenated version (e.g. -online.com).

So I updated my response to the question to reflect this.

Sebastian

I don't think it maters to Google, but it does maters to the person searching. I think that using a hyphenated domain increases the readability of the URL and therefore leads to better CTRs (and thus to a tiny advantage in rankings).
Also if someone misspells your brand name it's easier to scan in the SERPS what you were looking for (e.g. searching for "brant product" leads to brand-global.com, which I think is a lot easier to read then brandglobal.com)

Personat opinion, withouth any data: most ppl. tend to use hyphens when typing in a name.[edited]

Low hanging fruits are usually common errors/mistakes someone made. So for starters I would do the following thing:

Register with Google's webmaster tools.

Crawl your site with xenu's link sleuth (google it, its freeware).

• Look for 404s -> fix them
• Have a look at all titles of  you page: are they unique, short and do they have the important keyword in the beginning.
• Look at the depth of your page. Anything above 4 should be looked at.
• See whether all pages send the right status code (404, 200) and the right charset
• Analyze one page with Google's Pagespeed Browser Plugin, fix whatever comes up

So you want to get a few long tail querrys and then have the visitors buy something? My first guess would be to use the first approach and put them in the article section. Put a nice big button "Buy some nice jewlery made with [technique]" in the beginning and end of the articel.

I think the main advantage is that you can crosslink from every page that contains these kind of products to the page "Read how they are made here" (think wikipedia) and you can expand this section easy: going from a general overview of materials to detailed pages containing special information ect.

Key is to get the visitors from those pages to the products and jewelers.

Answered to a similary question :

I see it this way: if the topic is about your webpage and you can add something valuable to the discussion then go for it, but see it as a way to build an audience and not so much links.

Even if the links are nofollowed or not indexed at all: nofollow-links are essential to a "natural" link profile, as most websites do have a number of nofollowed links (and even links from noindexed pages, which google crawls and "sees" nevertheless).

I see it this way: if the topic is about your webpage and you can add something valuable to the discussion then go for it, but see it as a way to build an audience and not so much links.

Even if the links are nofollowed or not indexed at all: nofollow-links are essential to a "natural" link profile, as most websites do have a number of nofollowed links (and even links from noindexed pages, which google crawls and "sees" nevertheless).

Ask yourself: What structure offers a better user experience? Which one is easier to read, more concise, offers more orientation? Use that one!

Do you really need folder-folder-page ? How about just ".com/page.html" ? Or at most .com/category/page.html

I am pretty sure that

widgets.com/green-widgets.html has a better CTR than

www.widgets.com/widgets/color/great-green-widgets.html

You can also use the really cool freeware tool "Xenus Link Sleuth" and crawl your own page. It has a lot more featrues to offer like page depth, header information etc etc. and best of all: it is free

Just crawl your site and sort by status. Then rightclick on the 404ed page and see what pages link to that page.