David Leonhardt’s SEO and Social Media Marketing

A while back, I wrote about why to ignore three-way link requests.  Many of the reasons I listed had to do with the quality of the site linking back to you.  But what if it’s a PR3 home page.  Sounds like a juicy link to score, doesn’t it?  Well, maybe not.  I don’t want to pick on one domain or another, but I need an example, so the one that came in today will do.  In the words of the link-exchanger:

Mate its PR 3 schoolsprepared.org

 

Check it again..not throwaway…

There are so many domains like this, and while a link from that page might not carry zero value, it’s caveat emptor.  Here are seven reasons why this is not a ” Wow! A PR3 home-page link!”

The domain will get dumped.Like so many others, this domain used to be a real website, but no longer. One glance at it with naked eyes shows that it was nicely set up and had a purpose. It accumulated a PageRank of 3, which means it was somewhat active on the Internet. And like so many others, the owners bailed out and sold the domain to someone who thought a PR3 website would be great for three-way link exchanges. So what happens once the site is “used up”? Once it is so stuffed with links that it is no longer useful for attracting link-exchanges, what do you think will happen to that website (and your link on it)? Come on, be honest, do you really trust that they will continue to maintain the website?

The page will fail to keep up. Let’s suppose they do maintain the website, honestly remaining committed to protecting the link they posted to your website, as promised. How long will the page remain PR3. Remember, PageRank is relative; as the total number of web pages and the total number of links on the Internet increase, so too does the link juice required to maintain a given PageRank. But the owners are not building links to this site; they are building links to another site.

The page will not attract new links. The eyeball test tells you this is a link farm. Even if it isn’t technically a link farm, it looks like one on first glance. Nobody will want to link to it. No bloggers. No industry sites. Nobody. The owner could be less careless and format the links nicely. But, as with most such situations, the owners did not.

The page will suffer link attrition. OK, let’s take this one step further. Over time, all websites suffer from link-attrition. That is to say, links die every day (websites close down, links pages are cleaned up, links get pushed deeper and deeper on directory pages, etc.), and links pointing to the page your link is on will die. In the case of a website that looks cheap like this, it stand to suffer accelerated attrition, as some websites linking to it will remove their links when they realize what they are now linking to.

No targeted traffic. As Yura Filimonov pointed out to me, sites like this won’t deliver targeted traffic.  Anyone who lands on such a page will quickly see that it is useless and back out the door.  Of course many links don’t deliver much traffic, but one of the benefits expected from a home page link is some targeted traffic.

PageRank will be diluted. Eventually there will be dozens, maybe hundreds of links on the page. The PR from PR3 (what’s left of it) will be diluted before the domain gets recycled, is dumped or simply disappears.

You are not fooling the search engines. If I can see with a glance that this is a flipped website turned link farm, do you really believe that Google and Yahoo are being fooled? Please, don’t flatter me; I know they are smarter than I am.

“So, OK, David…would my link on a page like this place my website at risk?” you ask.

I doubt it.  If you have 100 inbound links and 80 of them are from home page link farms, that might throw up a pretty big red flag.  But if you have a dozen links on silly pages like this amongst 500 links of various quality, I can’t imagine it harming your rankings.  Just don’t go jumping for joy thinking you’ve struck gold.  You’ve just found a penny.

Related reading on a humerous note: a spammer link exchange note.

“So if I understand correctly,
you want to offer me a minor league bench-warmer
in exchange for one of my star players?”

Like you, we are inundated with link requests. Some are worthwhile; some are not. Obviously if the topics are way off base, we don’t pay any further attention. If the topic is on-base, we give the site a quick eyeball test.

But what do you do when a three-way link is requested? Just to understand correctly, a three way link is when a webmaster or hired gun requests a link from your site to their site, and offer in exchange a link from another website altogether.

Here are nine solid reasons to ignore such emails…

There is more work up front, because there are two websites to evaluate for one linking opportunity. Given the relatively few worthwhile requests, it really is not worth having to work twice as hard to decide whether to pay further attention, when you can work many times less hard by just deleting.

As a general rule, three-way linking requests are on average of a lower quality than straight two-way swaps. So there is even more incentive to delete the message rather than work twice as hard to review it.

Most three-way link requests want you to give a link from your high-quality website (your star player), but offer you a link from their low-quality website (their bench-warmer). If they are not interested in linking back from their quality site, why should you even bother?

The website they are promoting is the one they want a link to, naturally. Over time, as they build ever more links, a link from that site will become valuable. They offer a link from a site they are not promoting, which through attrition will become less valuable. Yet they are asking you to link to them from the site you are promoting and making more valuable. Fair? I think not.

Very often the website they want to link to you from is not even theirs. Very often, it belongs to a linking specialist they have hired. The linking website in most of those cases exists solely to provide reciprocal links, making is not only poor quality, but also expendable; when the linking campaign is over, the hired guns have no interest in keeping your link, poor quality as it is, live.

With alarming regularity, three-way link requests come in offering return links from the same website, often a directory. Link to Site A and get a link back from PainInTheButtDirctory.com. The next day, link to Site B and get a link back from PainInTheButtDirctory.com. The next day, link to Site C and get a link back from PainInTheButtDirctory.com. The next day, link to Site D and get a link back from PainInTheButtDirctory.com. It’s actually kind of amusing in an annoying, make-them-stop sort of way.

Contrary to popular opinion, a three-way link does not fool Google and Yahoo, at least not if it is done on any measurable scale. If a website has a high percentage of inbound links from websites that are all linked to from the same website or family of websites…well, let’s just say that the one thing huge data processors are expert at is recognizing patterns. ‘nough said.

Do the search engines value three-way links more than two way links? Some people swear by it. I have no data to back up the assumption I am about to share with you, and I could be very wrong. However, I suspect that the search engines see two-way link swaps as both a means of boosting link popularity sometimes and as a means of partnership marketing sometimes. What percentage probability they attribute to which characteristic is likely based on features of the linking pages. However, when there is a three-way linking pattern detected, I am pretty sure the search engines would attribute close to 100% probability that the links are solely to manipulate their ranking of your website. When else do you ever hear of three-way linking? I suspect that these links are seen as fairly dark gray-hat SEO, if not black hat.

The very last place you want to be found in is in a link directory that is solely used for three-way links. It’s like waving a big red flag and calling out, “Hey there, Google. Yoohoo, Yahoo. Lookit me. I’m messing with your rankings.” Is that the message you want to send the search engines about your website.

Are there legitimate three-way link requests? I have to add this footnote, because there is nothing black and white, not even SEO hats. There are some rare situations where a brand new website approaches us with a link request. They know that their links page carries no weight and might not even be indexed or cached by the search engines. In other words, they know they have nothing of value to offer in return. So they offer a link from another website. This happens rarely, but when it is specified that this is the reason, I usually take the time to look at their website and consider their offer. And, assuming the site seems worthwhile, I sometimes ask for a link back from their “worthless” page, knowing that over time it will be worthwhile…and in many cases more so than the third-party link page that seems more worthwhile in the short term.

I’ve been tagged in a meme.  I think I was tagged sometime a ago in one, but this time I’m paying attention for a few reasons:

1. It’s a good chance to tell you about the linking benefits of participating in memes.
2. I was tagged by SEOAly (Alysson Fergusson), the Sunshine of Twitter
3. SEOAly might get to go free to the IM Spring Break conference (and surely bring back some great notes for me, since I don’t have time to attend)

The concept of a meme is quite simple.  At heart it’s a tag-you’re-it game, where one blogger “tags” several others by linking to them.  Their role then is to link back to the person who tagged them, and then tag additional bloggers.  If you can visualize the linking diagram, you can see the benefits this brings to your blog and to the blogs of your friends and associates. 

If a large part of SEO is making your website well-connected across the Internet, a meme helps establish the community of like websites.

So to participate in this meme, I am supposed to link back to SEOAly, link to the IM Spring Break meme page, list 7 things you expect to happen at IM Spring Break, then tag as many people as possible.  (I will cheat on that last one; I’ll tag just a few people who have been readers of this blog whom I think might not yet have been tagged by someone else.)

What will happen at IM Spring Break.

1. I will miss 100% of the sessions, because I’ll be back home working for my clients.
2. I will miss Chris Winfield’s insightful presentation, but I will get to contribute to it when he seeks input via Twitter as he did for his Lawyers on Twitter presentation last month.
3. Several words that I can’t repeat here, in case my kids ever read this, will punctuate an equally insightful presentation by Sugarrae Hoffman.
4. Jordan Kasteler aka Utah SEO Pro will get even with the IM Spring Break organizers by going 100% dry all conference.
5. SEOAly will attend, thanks in part to this blog post.
6. SEOAly will bring back notes for everyone she tagged.
7. I will run out of ideas before I get to #7.

So, let’s tag Wilson Raynopss, Peter Lee, Mr Javo, Communibus and Barry Welford just for fun.

Is a meme a good way for you to spread the word, too?

Do you trade links with PR0 pages?  Once upon a time I avoided PR0 pages.  It was usually a sign that a web page was being penalized or suffering from some contagious tropical disease.

But times have changed, and my approach has changed with the times.  In recent months I have seen a lot of “links” pages with PR0 value on the Google Toolbar.  This includes pages that are linked to from every page of otherwise PR4-PR5 websites.  PR0 in the Google Toolbar is n o longer, in my opinion, a kiss of death.

If I am offered a link from a PR0 page, the first thing I do is give it the old eyeball test.  If it has a lot of the wrong kind of links and the wrong kind of spammy words all over it, that ends it there.  But if the page looks good (on topic, manually-maintained, etc.)  apart from the lack of a green bar, I take a look at the home page of the site to see if it has a green bar and to see what the path is to the link page.  Is it linked to from the home page, from the template, from a second level page?  

In other words, I’ll make my own call at roughly what the value of the page is.  And given the error-prone toolbar that is at best an approximation anyway, I am sure my calculation isn’t that far off.  

Is this process more work than just looking at the toolbar?  Yes.  But when the toolbar is blank, the only alternative to this process is to just ignore what could be a good link.  Given how hard we work to find good, on-topic links, I think the work is worth it.

You can easily tweet this post by clicking reTWEET this

Three months ago, I wrote about how offline links count, too.  I recounted the story about how hard it was for me to enter a contest, because of the typo in the URL on a printed flyer.  Well, it seems the contest is on again.  Yesterday, I received a new flyer in the mail from the Lake Placid folks with a new entry code and…the correct URL.

How will this new flyer by viewed by people?  We can only speculate, but here are some possibilities.

• People who did not try to enter before will most likely see this as brand new, so the gaffe would not affect them.
• Some people who did not try to enter last time might remember the previous flyer, in which case it served as branding and might increase the person’s likelihood of entering this time.
• People who tried to enter last time might try again.
• People who entered last time might figure it’s a waste of time to try this time.
• Many people might not try to enter the contest, but they will get the message to visit upstate New York, which is the whole point of the contest.

There will probably be plenty of people who fall into each of these categories, and perhaps some others, too.  Hopefully for the resorts involved in the mailing, most of it will be positive.

And I’ll bet they spend more money on proofreading in the future!

Dear reader, let me be a heretic once more.

We all know, or at least assume, that having multiple links to the same URL from a domain is an exercise in diminishing returns as far as search engine rankings are concerned.  That is to say, if you score a link to your home page from one page on a domain, any additional links to your home page from other pages on that same domain are worth less.  And the more links to your home page from that domain, the less each one is worth.

This makes sense.  If a domain has 1000 pages, a sitewide link cannot be viewed as 1000 endorsements for your home page.

But the web is a changing place, and in the past few months, services have been cropping up to submit your website to 1000 and even 2000 social bookmarking websites.  These services are similar to all those directory submission services and the article submission services, and they are often offered by the same people.  On the surface of it, there is nothing wrong, but it does require a reaction from the search engines.

But first, a personal rant.  Submitting your home page to 2000 social bookmarking sites is NOT social bookmarking.  It is bookmarking, but it is NOT social.  If it was social, these services would be promoting your page on these sites, networking with other users, and you would end up with several links at any one social bookmarking site (assuming your content is actually interesting).

OK, that was more than just a personal rant.*  In fact, I’ll bet the search engines are noticing the same thing and looking at the same numbers and raising one of their search engine eyebrows right now.  If there are thousands of single-link entries at each social bookmarking website, most of which are essentially paid links, should those each be worth more than each entry that garnered, let’s say 12 Diggs or Zooms?  Those dozen votes clearly are exactly the type of recommendations the search engines look for in their algorithms.  Single links at social bookmarking websites clearly are not.  Each Digg or Zoom should be worth more than each single entry.  In fact, we might even go so far as to say that the more Diggs or Zooms, the more each one should be worth.

What should the search engines do?  Clearly, their algorithms must distinguish between sitewide links and links that appear numerous times independently on the same website.  This is true not just for social bookmarking sites, but also for forums where a resource might be cited in numerous threads over time.

Maybe Google and Yahoo and MSN already do this.  Maybe I’m not being that heretical after all.  Naw, that just would be too out-of-character.

* It qualifies as a rant because I capitalized the “NOT”.  Twice.

Matt Cutts, Google’s public face for webmasters and search engine consultants, has shown us how to do link bait.  Oops, I mean, how to do really good quality content.  Yeah, that’s what I meant to say.

Here is the link bait…I mean content:

http://www.mattcutts.com/blog/9-google-mobile-iphone-tips/ 

Note that it is a numbered list, and not a “top 10″ list.  Matt chose a top 9 list, which is just a little offbeat..  Note that there are plenty of illustrations.  And the text and images combined are useful - actually demonstrating how to do something - not just silly stuff (although sometimes I like silly stuff, too).

Matt submitted it to Digg: 

http://digg.com/apple/9_Tips_for_Google_s_New_Voice_Recognition_App_for_iPhone

As of now, it has 42 Diggs. 

Study it hard, becasue even if your content doesn’t get more than one or two Diggs, this is how the Google guru prepares his content, so you can’t go wrong posting something like this on your website. 

There now, Matt just got a link from me as a result of his quality content.  You see?  It works. 

I’ve been meaning to blog about your desperate need for sucky links for some time, because I have not seen this aspect of link quality discussed anywhere.

People approach me all the time asking for high-quality links.  Not surprising - who would want low quality links?  But if you ask an SEO consultant to build you only PR6+ links, consider what message that sends to the search engines.

At worst, Google will assume you are buying links to buy PageRank…and we all know how much Google loves link-buying to boost PageRank, don’t we.

At best, the search engines will think your site appeals only to some kind of an elite.  How else would you explain that only high PageRank (high traffic, high-trust, etc.) pages link to your website?  Why do smaller blogs not link there?  How come your website is not included in any normal directories?  Why does this website have no appeal to normal people…and why should we rank it if it has no popular appeal?

No, the search engines won’t ask these questions outright.  But remember that all algorithms are created to simulate what would be normal linking and trust patterns that real people would follow.  Having links only form high quality, top-ranked websites does not look normal.  It’s a giant red flag.

Ironically, the more high-quality links you have, the more poor quality links you need.

CAUTION:
By “poor quality”, I do not mean spammy websites.
I do not mean you should be on pages full of words related to
enhancing body parts and gambling away the kids’ inheritance. 

But I do mean, that you want links form websites with a variety of linking profiles, ones that might be new or might not be running any link-building campaigns, ones that we might consider much less significant.  In short, you want a normal linking pattern.

The ideal way to build links is still the tried and true…

Step one, create awesome content such as useful articles, instructional videos, samples and demos, all the things that are generally called “link bait”.

Step two, publicize this content.  If it really is good, many websites will link to it, including top-rated websites and many smaller less significant websites.  They will do it naturally. So you will have a natural linking pattern.

To answer the obvious question, yes you will surely want to put extra effort into publicizing your content to high-trust, authoritative websites.  But those links are the kind of links that less-significant website owners will follow, read and link to, as well.

So don’t forget to get links form a wide variety of insignificant websites as part of your link-building campaign.  With algorithms designed to simulate something like democracy, votes from “the little guys” count, too.

This is a tough one.  I have a habit of trying to fit as much into a title tag as possible.  I try to keep it below 12 words as a general guideline, but I know there is a difference between a short 12-word title, such as “Dine in - fast food fair for people who eat on the go” (53 characters) and a long 12-word title, such as “California pipeline producers - environmentally sensitive petroleum infrastructure for transporting energy internationally” (OK, so it’s just 11 words, but it’s 122 characters).

In addition to being the most valuable real estate on a web page, the title tag is also used by many other websites to link to your page.  So someone linking to the pipeline site, might use “California pipeline producers - environmentally sensitive petroleum infrastructure for transporting energy internationally” as the link text.

A new study suggests that Google reads only the first 55 characters of link text, which means that in the above example it would read only “California pipeline producers - environmentally sensiti”.  If you had to target such ridiculously long words for your search market, that would totally suck.  But it does speak to the importance of placing your most important keywords at the beginning of your title tag.  This is where so many websites that put a corporate name of even their domain name at the beginning miss the mark.  It also means trying to keep words like “the” and “and” out of those first 55 characters. 

A couple caveats: Google could at any time change this to 50 characters.  Or 60.  Or 600.  Or 10, for that matter.  So don’t get stuck on the number 55, but focus on the principle.  Yahoo and MSN will have their own limits, too, so don’t stop at 55 if you have something that might get picked up by someone else.  And, of course, the title tag is not primarily about link text, is it? 

This will be a short post (I hope!).  Just a few days ago I returned from some fun in New England, and I was thinking about returning into the mountains of New Hampshire or upstate New York.

So it was with more interest than usual (I didn’t rip it up) that I opened a flyer that read”VOUS venez peut-être de GAGNER un des six forfaits escapade à Lake Placid”.  This delightful and hopeful notice was followed by a website address where I could enter a code to verify that I had indeed won a package to lake Placid: www.LakePlacid/WIN .

So I did.

Or, at least, I tried.

I tried again. 

And again. 

I tried adding .php and .html and .asp … all to no avail.  many people would have given up at this point, or much earlier, but I have a stubborn streak that you really don’t want to catch from me, and finally I figured out what was wrong; they had forgotten to print the .com in the URL.  I knew what to look for, and still it took me a while.  How many people would not have known that a URL is invalid without a TLD?  How many people would have given up without even trying to fix the URL? And most importantly, how much money did the Lake Placid Essex County Visitors Bureau invest in designing and mailing these brochures that were missing four crucial characters?  It is a mistake I am sure they will not let happen again.

When building links, one of the points that even the legions of outsourcing link-builders in India will focus on is that they will make sure to post the correct URL without typos.  Your offline link-building is just as crucial.  In fact, a typo in one online link doesn’t matter too much.  A typo in a pamphlet that hundreds or thousands of people will read matters much more.

By the way, I did not win that package…but I should get an A for effort.  And if nobody else figured out the correct URL, maybe the Lake Placid Essex County Visitors Bureau will award me the prize by default and I’ll get to do some 46er trekking.