Central Perk

marți, 23 august 2011

SEOmoz Daily SEO Blog

Linkscape Updates and Wireframes/Feedback on New Task List PRO Feature
What's the Toughest SEO Gig You've Ever Had? I'll Start.
Data: Identify Yourself Authoritatively for More Followers

Linkscape Updates and Wireframes/Feedback on New Task List PRO Feature

Posted: 23 Aug 2011 06:39 AM PDT

As you may have seen, Linkscape has just updated with a brand new index crawled primarily in early August. This means there's new data in the web app, in Open Site Explorer, the Link Intersect tool and the mozBar (note: the new Firefox 6 caused a bunch of issues w/ our toolbar displays, so it will be another week or so before it's re-compatible with this version).

In Linkscape's previous index, we focused a bit more on crawling large, powerful, important domains more deeply at the expense of some smaller, low mozRank sites on the fringes of the web. This time, we're trying to compromise a bit with an index that's a bit more balanced - it's still quite large; 51 billion URLs (vs. prior index sizes in the 38-42 billion URL range), but contains more unique root domains - 98 million vs. 91 million in the prior index.

As always, we'd love your feedback on this index's quality and value for competitive analysis / link building. Oh, and BTW, for those of you interested in link building with Linkscape data / Open Site Explorer, I'm giving a webinar on that topic this Friday, so sign up!

Here are the latest Linkscape stats:

51,494,511,857 (51.5 billion) URLs
592,299,695 (592 million) Subdomains
98,626,331 (91.6 million) Root Domains
399,327,725,405 (399 billion) Links
Followed vs. Nofollowed
- 2.25% of all links found were nofollowed
- 58.82% of nofollowed links are internal, 41.18% are external
Rel Canonical - 10.12% of all pages now employ a rel=canonical tag
The average page has 78.03 links on it
- 65.09 internal links on average
- 13.21 external links on average

I've only run a few reports so far, but my sense is that there's a reasonably good balance here, though, of course, I'd still love to get a larger index and hope to achieve even more significant growth on that front over the next few months. If you ever find a link we're missing or something you expected that we didn't have, please let us know.

I also have an upcoming feature of PRO that I can share today, and I'm pretty excited about it. The feature is currently called "task list" and it's part of how we're planning to make the SEO process and the data inside campaigns more actionable and useful. Here's a wireframe (note: not a screenshot!):

PRO Task List Wireframe

The idea is to give marketing professionals working inside the app a list of potential "to-do" items. Obviously, we may not always know what's on your plate, but from the weekly crawl, ranking analyses and on-page reports, we can often get a sense of some good priorities. Along with these, we want to add a list of common tasks that every SEO may want to engage in with a campaign/site. You can see some examples of that in the wireframe above, but we'd also love to get your feedback/suggestions.

To that end, we've got a survey on the task list you engage in as a professional marketer, and we'd love to get your input here.

Thanks a ton - and feel free to add any color or additional ideas in the comments below!

p.s. One more item - the anchor text tab in Open Site Explorer will be showing data from last index for a couple weeks, as we encountered some processing issues of anchor text in this index. We hope to have that fixed soon.

Do you like this post? Yes No

What's the Toughest SEO Gig You've Ever Had? I'll Start.

Posted: 22 Aug 2011 01:30 PM PDT

Posted by randfish

I recently spent a few days in New York, Philadelphia and New Jersey, and got to meet with great friends at consulting firms. One conversation in particular stuck with me regarding the hardest project they'd worked on in the last few years. It got me thinking that I should share a few of my most challenging SEO projects, and see if the community had stories they could share, too. I learned a lot in my short chat on the subject, and I'm hopeful that will translate here on the web, too.

Thus, my top 3 "toughest" SEO projects over the last decade were:

The "I Just Want PageRank 10"
It's hard to believe in 2011, but in 2005, Moz really did have a client who's entire goal was increasing toolbar PageRank. He was a very wealthy individual who fretted over his PageRank score in comparison to a rival's organization in the same sphere. We were able to grow it from 6 to 8 with mostly white hat stuff (yes, even I used to dabble with link buying back in the day), but when he continued pressing for a PageRank 9, we gave up and sent him to a link broker. We provided very stern, very aggressive warnings that this could hurt his traffic or even make Google penalize the PageRank score, but he wanted all the options and did end up spending a small fortune on paid links. To my knowledge, the PageRank score never went above 8/10, and of course, since he wasn't doing any keyword targeting to speak of and didn't care about traffic, the value of those links was virtually nil.
_
The "Breaking Out of the Sandbox"
One of my first few SEO consulting clients, starting in 2003, was a commercial lender in Seattle. I did very old school SEO for them (not knowing any better, and really learning the practice as I went along), jamming anchor text and links wherever I could. Unfortunately, this was right around the time of Google's "sandbox effect," which a Googler explained to me once as "an artifact of something else we created." For nearly 18 months, I fought against the sandbox, earning #1 rankings for nearly every term in MSN search and Yahoo! (which, at the time, had much more significant market share), but page 10+ on Google. Then, suddenly, one morning, we were #1 across the board in Google, along with a number of other sites that also "popped out." One odd element of the sandbox was that Google seemed to release sites en masse, similar to the Panda updates this year. We celebrated heavily, the company started getting customers and over the next few years, we paid off a lot of nasty debt.

But, I remember most that every day and every night, I did virtually nothing but crawl the web (manually - there weren't any tools like OSE or Link Intersect) and try to find link opportunities of any kind. I hadn't really discover the power of inbound marketing, and the social web was nearly non-existant outside of blogs + forums, which I used heavily. Walk to work. Build links. Walk home. Think about other places to get links. Eat dinner in a crappy, old 500 sq ft. apt. with Geraldine. Build links until 2am. Get up. Do it again. I sometimes wonder if we hadn't broken out whether I'd have career at all today.
The "Linkbait that Brings Customers"
Later in my career - around 2007 - we had a client that wanted to invest heavily in linkbait, but wasn't quite sold on how the process worked. To be fair, neither were we. We'd seen plenty of examples of linkbait being very successful in driving links, which then drove great search rankings and more high quality traffic (vs. the typically more high-bounce-rate visits to linkbait pieces) and on occasion, we'd had success with a linkbait piece that did get email addresses or trials of a product. But, this was in the housing/real estate field, and those consumers are very research-heavy. It's hard to get a first-time visitor from any channel to fill out an application or leadgen form in that world, but this was the project, so we executed the best we could.

Needless to say, it failed. Of 3 linkbait content attempts I recall, 2 generated some decent links and 1 was a real success, but only from a links perspective. The customers never appeared, and the client left, unsatisfied. If I recall properly, we refunded half of the cost and lost money on the deal overall (we hadn't charged enough in the first place, honestly), but wanted to preserve a positive relationship. Lesson learned - create realistic expectations and don't agree to a project that goes against the laws of marketing.

Of course, there have been (and still are, though I only provide consulting pro bono and through Q+A, now) plenty of other big challenges, but these ones stand out in my mind. I also asked some folks on Twitter and Google+ and received some terrific responses, highlighted below:

You can read the Google+ thread here and see Twitter replies here (at least for the next hour or two). I'd love to continue the conversation and the sharing in the comments. I'm flying back to Seattle tomorrow, but will try to be in the comments. I've also got a post coming tomorrow AM on a new Linkscape update.

Do you like this post? Yes No

Data: Identify Yourself Authoritatively for More Followers

Posted: 22 Aug 2011 10:25 AM PDT

Posted by danzarrella

This post was originally in YOUmoz, and was promoted to the main blog because it provides great value and interest to our community. The author's views are entirely his or her own and may not reflect the views of SEOmoz, Inc.

For more data on social media and busting of unicorns-and-rainbows social media myths, be sure to register for the Science of Social Media webinar that is being held on Tuesday, August 23rd at 2:00pm EST. This webinar is actually going to be certifed as the largest online marketing seminar ever by the Guinness World Records folks.

The easiest social media myth to bust is “don’t call yourself a guru.” Proponents of this myth argue that self labeling yourself as an expert makes you sound pretentious. Maybe it does, maybe it doesn’t. But the data clearly shows that telling your audience why they should listen to you absolutely works to increase your reach.

When I first started analyzing Twitter account data, one of the first things I noticed is that a surprising number of accounts don’t include profile pictures, bios or homepage links. But when I looked at the number of followers accounts with and without those things have on average, I found huge differences.

It may sound pretty obvious, but users who’ve taken the time to identify themselves with a bio, picture and link tend to have many more followers than those who haven’t taken that time. The above graph shows the effect of including a photo, but the effect is the same with bios and links. In real life and in social media, if I know who you are, I’m much more likely to listen to what you have to say.

Taking the identify yourself suggestion a step further, I analyzed common words found in Twitter bios. I found that users who included authoritative words like “official,” “founder,” and the dreaded “guru” tended to have more followers than the average Twitter user.

Once I know who you are, if you’re someone important I’m even more likely to want to hear what you have to say. If you’ve founded a company or authored a book, I’m interested in what you think.

Social media isn’t that different from the offline world. Introduce yourself, tell us who you are and why we should listen to you.

Do you like this post? Yes No

Seth's Blog : Two earthquake-related thoughts about human nature

Two earthquake-related thoughts about human nature

1. The first thing that happens after we encounter an earthquake is to wonder if anyone else felt it. The need for group validation is widespread and happens for events that don't involve earthquakes as well.

If those in the tribe feel something, we're likely to as well. That's why people look around before they stand up to offer an ovation at the end of a concert. Why should it matter if any of these strangers felt the way you did about the event? Because it does. A lot. Social proof matters.

2. Organizations are busy evacuating buildings, even national monuments. Even though experience indicates that the most dangerous thing you can do is have tens of thousands of people run down the stairs, cram into the elevators and stand in the streets, we do it anyway. Why? Because people like to do something. Action, even ineffective action, is something societies seek out during times of uncertainty.

• Email to a friend •

President Obama on Libya: The Future is in the Hands of its People

Your Daily Snapshot for
Tuesday, August 23, 2011

President Obama on Libya: The Future is in the Hands of its People

Yesterday afternoon, following a call with the National Security Council, President Barack Obama spoke about the evolving situation in Libya. Over the past six months, the United States has worked with allies to protect the people of Libya from Muammar Qaddafi's brutality and support them as they seek the opportunity for the citizens of Libya to determine their own destiny.

President Obama said, "The Qaddafi regime is coming to an end, and the future of Libya is in the hands of its people," making it clear that the courage of the Libyan people has brought freedom within reach.

Read the full statement or watch the video here.

Reflected in a mirror, President Barack Obama conducts a conference call with his national security staff, including John Brennan (right), Assistant to the President for Counterterrorism, on the situation in Libya in Chilmark, Mass., August 22, 2011. (Official White House Photo by Pete Souza)

SEOptimise

Experiment: Do Google +1s Impact Your Rankings?

Posted: 22 Aug 2011 01:41 AM PDT

Several months ago I ran some tests on the impact of Facebook Likes on Google Rankings. After receiving several e-mails and tweets asking whether Google +1s have any impact on rankings, I decided to run some similar experiments in order to give a more conclusive answer.

Note: on September 19th I'm going to be presenting a few insights on social SEO at Social Media Marketing and Monitoring 2011 London. If you're interested in learning more about Google +1s and how to harness the benefits of social signals, come and visit! I've included discount codes and links to the conference tickets at the bottom of this post.

I began my Google +1 experiment by taking 35 URLs across six unique domains with various traits: a mixture of indexed/unindexed, had links/no links, home pages/deep pages, had social signals/no social signals etc. All of these domains had experienced no significant changes in rankings for a while, i.e. the main keywords that I was tracking had not moved for a fair amount of time and there was very little other activity being done on them.

I then requested some +1s from a variety of different groups of people and monitored the log files, rankings, Google Analytics, and Webmaster Tools' Google +1 section to spot any interesting trends.
Results:

	Type of page	Indexed?	# of +1s	# of links (internal / external)	Ranking observation
URL 1	Deep Page	No	20	0 / 0	No change
URL 2	Home Page	Yes	26	12 / 1,136	No change
URL 3	Deep Page	Yes	20	595 / 0	Ranking increased
URL 4	Home Page	Yes	26	0 / 3	Ranking increased
URL 5	Deep Page	Yes	28	307 / 30	Ranking increased
URL 6	Deep Pages	Yes	30 over 30 pages	N/A	Rankings increased

Analysis:
Although it has only been one week since starting the test, and the sample size (35 URLs across six domains) is not at all statistically significant enough to make any rash conclusions, my general observation is that +1s from genuine Google+ users do improve rankings. However, in the case of URL 1, which was the only URL that was not indexed, the +1s had no impact on getting it indexed.
The scale of some of the ranking increases was also very varied: one of the URLs increased the ranking for a targeted keyword from 75 up to 27 in Google UK. Another Keyword increased from 42 to 18, and another increased from 8 to 6.

Google+1s Impact on Traffic
Here is a Google Analytics screenshot for URL 6 (the domain where +1s were spread across 30 URLs). As you can see, after relatively stable traffic there does appear to be a slight overall increase in traffic – although it's hard to say whether this is was caused by the +1s or if it is just correlation. Either way, it hasn't smashed the site’s traffic, but it is a noticeable increase (from ~185 visits a day to ~225).

If you want to learn more about this experiment and how to harness the power of social & personalised search, I am going to be speaking at Social Media Marketing & Monitoring 2011 London on September 19th. Tickets are available now and you can get a 10% discount by using the following discount code: SPKR10.