David Leonhardt’s SEO and Social Media Marketing

If you plan to haggle over PageRank with me…goodbye.

That’s right, I have kicked the habit.  The size of your PageRank doesn’t impress me any more.  PageRank surely is still real, but an individual page’s PR can often shrink or grow so that neither you nor I can really know its real size.

• The Toolbar PageRank has always been at best an approximation. 
• Pages that show with PR3 or PR4 in the Google Directory are now often showing PR0 (PR Shrinkage)
• Whole sites are now showing PR0, even while they continue to rank as well as when their pages showed PR3 – PR4.
• Increasingly link-pages or directory-style pages are showing PR0, sometimes after showing PR3 or PR4 just a week earlier.
• The gray bar used to mean a page was not cached in Google – a sure sign of a penalty or a brand new page.  No longer.  Many pages with PR are now displaying the gray bar.
• Toolbar PageRank is dead!

Until very recently, I was assuming that the Toolbar was only showing false negatives – that if a page showed PR4 it was a pretty good bet that the PR of that page is at least PR4.  But recent observations have lead me to question this assumption, and perhaps I am jumping the gun, but I believe the toolbar is now showing false positives, too.

What I look for in a link exchange

Rather than PageRank, I look for a few other key items on the page where my client’s link will appear:

• Most importantly, I want to know the page is cached by Google.  Not only is that absolutely vital for the link counting with Planet Earth’s most important search engine, but it is a fairly good indicator of whether other search engines and real human beings will find the page, too.  Not cached?  I won’t even look at any other factors.  This is the show-stopper
• Is the page relevant to my topic?  If not, it had better be superb in every other area.
• Is the page relevant (optimized) to my search phrases?  Again, if not, it had better be superb in every other area.
• Is this page optimized for words like “link exchange” or “reciprocal links”?  Why not just type “SLEAZE” in big bold letters across the top of the page?  And don’t think the search engines can’t read words like “link exchange” or “reciprocal links”.  This is another factor that comes pretty close to being a show-stopper, too.
• Is the page part of some automated link machine script?  Let’s face it, you don’t want to send the search engines a message that, “Hey, I can’t get real links from real people who just love my site, so instead I found an automated script to keep me warm at night.”  This is usually a show-stopper, too.
• Once I see that a page is cached and passes the four eyeball tests above, it’s time to get critical.  The first thing I look for is a page that can easily be found.  If the page is one of 50 categories in a directory whose main page is linked only from the home page, that’s not a very good sign. Two clicks deep, and sharing link-juice with 50 categories?  I don’t think so.
• I also check that the page is not the last in a series of pages that link one from the other…that’s how many clicks deep from the home page?  Never go to dance with someone if she’ll make you stand in line to dance.
• I like a page that is either directly linked to from the home page or is linked to from a page that is in the sitewide template.  From an internal linking perspective, this tells the search engines that the page actually counts.  And you want your link to count.
• Of course, I also look at the quality of the website overall.  Is this a website that likely carries a lot of trust value?  Does it rank well for similar search terms to the ones I am targeting?  Is the link directory full of all sorts of totally unrelated categories, perhaps some of them even unsavory?  If this website sleeps around too much, be careful what you might catch.
• Is the page a content page?  I can forgive a number of other items for a genuine link in the midst of a page of text.
• Is my link last on a page with 500 links?  I really prefer pages with 50 or fewer links, but if there are more, I am fine with having our link added near the top of the list, but not at the bottom.

Tell me you have a high-trust website and a linking page that is well linked internally, relevant to my search phrases and clean from the flotsam that shouts out “sleaze”.  But don’t tell me your page has bigger PageRank than mine.  It’s not the size that counts; it’s how you use it.

For some time I have been pretty much ignoring PageRank in the Google Toolbar.  I know too many sites that lost big PR on certain pages and not others or lost it across the board, all with no noticeable affect in their rankings. 

More and more I see that link pages on websites register PR 0 (solid white bar) or no PR whatsoever (solid gray bar, which used to mean a site with a penalty) but which I can see by the PR of the rest of the website and the link structure should be all countr by PR2 or PR3 if not more.

But I have been assuming that the Toolbar shows PR lower than reality, never higher.  That is, it gives lots (and lots and lots and lots) of false negatives, but never any false positives.  However, lately my faith has been shaken.  There have been a couple offers of link swaps involving pages that just intuitively should not have such a high PageRank.  Today one of them struck me as odd enough, that I though I would blog about it.

This is a home page of a website that, according to both Google and Yahoo has 2 pages and shows less than 300 backlinks at Google.  Furthermore, it is a blog with just two posts, both from 5 days ago.  How would it get to be PR5, which takes a lot more links than it once did?  And why am I suddenly getting an email for a home page link swap (becasue the savvy owner realises that he has something to capitalize on quickly before it turns to dust!)?

Don’t trust that green and white bar. 

So somebody needs to build links to help his search engine rankings, and has come up with a very precise list of exactly what he wants.  It includes 19 exact specifications, which perhaps he pulled from a handy article somewhere on the Web.  Here is the list he presented, but this post could be abut any such list… 

1. One way non-reciprocal links only, no link exchanges.
2. THREE WAY Links where all links are in the same theme is OK
3. All links must be permanent.
4. Only 10% can be in directories.
5. No blogs
6. ONLY OUR THEME , (our theme is quite common so you will not have problems).
7. NO hidden links or any site that has hidden links.
8. No directories. No link farms, link-exchange programs, forums, Google banned site, black hat website. No guestbooks, links within forums, links within newsgroups or links from link exchanges etc. and never participate in any commercial web rings.
9. No sites banned by Google.
10. Link page must have a recent Google & Yahoo cache.
11. Must be manually submitted.
12. No Automated software (e.g., Zeus, Arelis or others)
13. All links must be from a different domain and IP address (geographically diverse, different class-c IP address block).
14. Only 10 to 15 links per week per language per site
15. Link pages must be static urls (no variables or parameters in the url)
16. No blacklisted or spam sites.
17. No more than 40 outbound links per page.
18. The link text must be from our keyword list and point to that keywords target page
19. All links must be static and without “nofollow” tags, no redirects, or javascript
20. Links must be on a PAGE with a Google PR of at least 2
21. All links must be on a page of the same language
22. Links must be on domains where we have no link

This post is about why I refuse to build links according to lists like this.  First, I must note that some of the items such as #9 and #7 and #15, for example, all make perfect sense.  These are deal-breakers that make a link useless. 

However, other elements are judgment calls: stipulating how many links per page, the PageRank, that a three-way link is acceptable but not a two-way link, among other factors.  What people hire me is to exercise that judgement.  To decode when a page might be PR1 but incredible on-topic and worth going after.  Or when a page might have 200 links, but with PR4 and lots of real human traffic it is worth its weight in gold.  Honestly, the client can just have his secretary or an offshore link-builder do the manual job of seeking out the links.  He does not need me for that.  What he needs me and my trained staff for is to exercise judgment – judgment that he is overriding with a pre-fab list. 

Does the client really think we have control over how many links are built in a week?  That depends on the response rate and the amount of back-and-forth with various webmasters.

And how much does he want to pay me to track down IP addresses to make sure they are all different?  Or check that the client does not already have a link on the domain?

That’s why I turn down offers to try to fit a strategic process into so comprehensive a list of technical specifications.

Today I just want to share with you Jason Lee Miller’s list of what works as good link bait and ideal for viral marketing.  His whole article is great and can be read here, but this is the list I thought I would share directly with readers. 

The Resource Approach (Becoming the Expert In Your Field/Niche)

–    Create expert articles/lists/data sheets 
–    Create practical or fun tools
–    Write How-To articles
–    Create a comprehensive blog roll (give link love, get link love)
–    Compile informative news stories and articles

The News Approach

–    Get the scoop. Be first with industry news
–    Interview prominent people in your field
–    Investigate a hot topic
–    Do an exposé

The Humor/Novelty Approach

–    Post funny/interesting/amazing photos related to your industry
–    Create humorous/unique videos (Use Blendtec for inspiration)
–    Create lists; people love lists – Top 10 Ways to…; 10 Signs You’re…

Most webmasters are at a total loss when they try to decide whether to do a link exchange.  In fact, they are so lost that they rely on how much green is showing on the notoriously inaccurate Google Toolbar.

Here is my top-5 list of how to decide if a link exchange is worthwhile.

1. The page is cached by Google.  That is the drop-dead bottom line.  If it is not cached, Google can’t find it.  And Google is the biggest search engine by far.  If Google can’t find it, chances are that Yahoo, Ask and MSN can’t either.  And chances are that real people won’t land on the page or navigate to it.

2. Relevance. The page should not be optimized for “links”. “link exchange” or “resources”, unless are searches you are targeting in your SEO efforts.

3. Relevance.  The page should be relevant for the specific words you are targeting.  In other words, the title tag and the heading should include at least one of the main words of the search you are targeting.

4. Relevance. The page should be on topic, regardless of specific words.  If it is full of totally unrelated websites, the search engines can see that it is just a collection of random links.

5. If you can get a link on a content page, or where yours is the only external link on the page, you have struck gold!

Let’s go hunt us a sacred cow today, OK?  Specifically, the sacred cow that all links must be reciprocated. 

There are a number of software programs you can purchase that can periodically make sure that all your link partners retain their links to your website so that you can notify them, warn them, threaten them and remove their links if they have removed yours.

Others spend time double-checking by hand.

Is this money well spent?  Is this time well spent?

No.

First, you have gained nothing by removing the links of those few who have reneged on their end of the bargain.  You have not increased your link popularity.  You have not gained additional PageRank.  You have not increased your website’s trust, relevance, content, number of pages or any other indicator that will lead to higher rankings.

Second, you have just spent money buying software that could have been spent elsewhere.  Or you have spent time checking backlinks that could have been spent creating them.

Third, you might even be doing yourself a disservice by making every outbound link on your resources page a reciprocated one.  The search engines are pretty clever.  They can detect when 100% of your outbound links are reciprocated.  They can detect when 100% of your links are part of a triangular linking pattern.  Do you think they are impressed with that?  My logic is that it is to your advantage if over time some of your link partners reneg and you have less of a pattern (remember that when it comes to linking, patterns should be avoided, for they indicate to the search engines that the links are contrived).

So, with apologies to everyone hocking link-checking software, my recommendation is to not waste another minute of the precious few God gave you checking up on your link partners.  A nice hike in the mountains would be a much better investment for your business…and of course for you!

I like this post saying Don’t Buy Rent Links . 

Julia Kristiva makes a great arguement for creating content, going through the costs and revenue, and how the website benefits from the content.  In this example, she references a useful tool created for a client.  But articles, data, case studies, ebooks, and other useful content can have the same effect.

I have never been a big fan of buying (“renting”, actually) links.  Just as with home ownership, it’s better to own than to rent.  If you own something that people want to link to, you effectively own the links.  But if the links are a result only of your monthly payment, your are renting. 

It’s the difference between a cost and an investment.

What do you do when someone asks you to do something that you are very good at, but just is not really worth their while?  Here is what I told a lady who wanted a news release about an interview she had posted on her website: 

It’s not that I mind taking your money…but I strongly suspect that a news release on this would be ignored.  That is not to say it is not interesting, but in your shoes I would allocate some time to contact specific blogs you think might be most interested about the interview and some of them will likely blog about it and link to the page on your site where it is posted.  Print publications are more likely to want to do their own interview. 

This approach will save you some money, although if you want to pay me for this advice, please fee free to do so by placing a link to my http://www.seo-writer.com website on yours. 

Have you ever noticed that some directories have a two-track navigation?  You can drill down to the appropriate category.  And you can also drill down to the apporpirate location. 

When submitting your website to such directories, always drill down as low as you can in the geography are.  Even if you are submitting a website that services the world, submit at the city level.  Even if the lowest level is not yet indexed by Google or yahoo or MSN. Here is why…

 Suppose you submit your website to http://www.wv-travel-directory.com/directory/0-world/187-cheap-air-travel/view.html .  You will get a link on that page.  If you take advantage of the deep linking opportunities (this directory allows three deep links, in addition to the home page URL), you will get 4 links all on the same page.

Suppose however, that you drill all the way down through http://www.wv-travel-directory.com/directory/10-north-america/187-cheap-air-travel/view.html , http://www.wv-travel-directory.com/directory/25-united-states/187-cheap-air-travel/view.html , http://www.wv-travel-directory.com/directory/316-new-york/187-cheap-air-travel/view.html , http://www.wv-travel-directory.com/directory/802-new-york-city/187-cheap-air-travel/view.html and submit your website at http://www.wv-travel-directory.com/directory/807-manhattan/187-cheap-air-travel/view.html .  I am, of course, assuming your offices are in Manhattan.  You get a link on the page you submitted to, plus on all the parent pages…all those pages you just navigated through.  Count them: 1 – 2 – 3 – 4 – 5 – 6.  Six links, and if you submitted three deep links for your listing, each of those links also shows up on six pages, so you get a total of 24 links.

Over at my favorite SEO forum, High Rankings, somebody asked for recommendations about SEO software.  The consensus response was, “Yes, don’t do it.”

But that response deserves a little more explanation, some of which was also posted at the forum.  There are two reasons why SEO software should be avoided.

Avoid SEO software for link-building.  What doe Google and company look for in links?  They are looking for recommendations.  They seek a sign that a web page is considered a good reference on a certain topic.  They are looking for natural links, not contrived ones meant to alter their results. 

What does automation do?  It creates patterns – patterns that are not natural, but contrived. What is the one strength that computers have that mere mortals like you and me and Paulina Rubio do not have?  The ability to sort through almost infinite data in almost no time at all and recognize patterns. Using SEO software is like posting a neon sign that reads, “Yoohoo! We’re trying to mess with your results.”

Think your SEO software can fool the Google algorithm?  Hmmm.

Avoid SEO software, because this is a sport.  I know a lot of web folks are techies who are used to the scientific principle that if you take certain steps, you will get certain results.  Repeat the same steps, get the same results.  SEO is not like this.  If a thousand people all repeat the same steps, there will not be 1000 websites in Google’s top 10 for “Paulina Rubio lyrics”.  There will still be just 10 results.

In any competition, the goal is not to duplicate what everybody else is doing.  The goal is to do more than everyone else.  To do better than everyone else.  And, if possible, to do what nobody else though of. It’s OK to study the competition.  It’s OK to study others who are not competitors.  It’s OK to take the best of what each of them is doing, but then you have to go out and do the very best you can.  No me-too software program will do that for you.

All that being said, I do use some software for SEO purposes. 

I use Internet Explorer to view websites (Yes, IE is software.)

I use Roboform to prefill forms for directory and article submissions.  But note that I manually edit important things like “title” and “description”.  Roboform just saves me from having to misspell my own name hundreds of times a day.

I use Keyword Discovery to help research the best search terms for my clients (It’s a web-based application, but it counts as software).

And of course, I use Word to compose articles and news releases, to edit source code and to do plenty of additional tasks.

And let’s not forget WordPress, which I use to blog about SEO. 

So software, yes.  Software to automate SEO, no.