Enrichment Depot – Profiting on the Net

Posted by wilreynolds

A few months ago at a keynote I asked people to embrace the HARD work of building a business [video]. We all fall in love with the concept of the four hour work week, don’t we…how can I make money and work less, right?  Who wouldn't. It’s natural.

The hard part is that almost every SEO I know who has a natural focus on scale has burned a more than their fair share of sites or clients in their career. There is just a mentality that comes along with the "scalable SEO", and it brings risks and baggage to many projects. I want to start getting people to think differently. If you can check out Bob Rains’ presentation from Distilled’s Search Love conference – (yes you have to pay, it's worth it) on why he’s turning white hat, you’ll get it. Scalable SEO is becoming HARDER work than you think and he's turning from Blackhat to White-ish.

Embracing the hard work, the time consuming, foundational building blocks needed to build almost any sustainable business (notice I am not saying a sustainable SEO strategy, I am talking about a sustainable business) is the key to long-term success. So copy this, print it out and tack this up in your cube…

There is no algorithmic update coming to correct rankings for tactics that are "doing hard work & connecting with customers".

Algorithmic updates are often targeted at those large-scale low quality SEO strategies, so lets just stop doing them. Let’s start with a historical view… 

PHASE I – It starts with cloaking…

Anyone around doing SEO pre-2000 did this, and especially scalable SEOs loved cloaking because they could create some page full of garbage text, and just insert keywords to hit a target keyword density, and boom page 1 of Excite, Lycos, or whatever.

Then came PageRank – Google realized that their search engine would be most successful if people had to "EARN it" by getting links from quality sources that had anchor text in those links. Google realized that too much money was at stake to just "trust" people to not keyword stuff, use white text on white backgrounds, and cloak (and they were right).

PHASE II – Scalable SEOs turned to reciprocal linking

Reciprocal linking sites sprung up quick to provide relief for the scalable SEOs hit by cloaking that was becoming less and less effective.

Reciprocal linking sites were not about finding sites to reciprocate with, who were quality, the strategy was all about scale…there were tools & scripts that were built PURELY to run reciprocal link exchanges. There is no added value or hard work associated with a script that lets anyone post a link to your "links page" and constantly checks to see if they have their link up to you still. Scalable SEO's loved it… run the script, get rankings. No real hard work required. Cash the checks.

Eventually Google updated the algorithm to hit sites who got most of their links in this scalable pattern. Leading to our next scalable exploitation… directory links.

PHASE III – Directory linking on a tragic scale

It wasn't long before scalable SEO's came in again this time with one-way links in the form of directories.

Awesome, now people are getting links in directories for rankings, but many of those directories don't really quality control the links either, pay to play. Remember, there are 230 quality directories for the taking.

As if that wasn't bad enough another cottage industry sprung up, you know the kinds of companies that will submit you to hundreds or thousands of directories for a very low cost.  This allows the 4 hour workweek loving SEOs to just pay some company to do all of their directory linking, sure clients might be at risk the day it stops working, but the scalable SEO doesn't worry about that, they'll just find the next exploit.  I believe that anchor text is going to be on the way out, maybe not in 2012, but soon I hope.

When that does happen, expect several sites to take hits because they have their eggs in too much of one basket, if directories are working for you now, start balancing you link portfolio.

PHASE IV – One-way bought links

Once directory links started working somewhat less (I believe they still work more than they should) next up were link networks. You know the ones that created the golden brick road for SEO's. Pay us, we put your one way, non directory links up on pretty quality sites. Instead of making legitimate relationships with bloggers, authors, site owners and journalists or doing things that are newsworthy, creating awards, badges, etc to get your links, you could just pay. So the scalable SEO once again rubs her hands together in utter joy…because creating relationships with bloggers is hard work, buying links on those blogs through a link broker is so much easier, all I have to do is charge enough, and get back to my 4 hour work week.

PHASE V – Panda smackdown, did you wake up??

I LOVE the fact that none of our existing clients were hit by panda, we (like many of you) picked up new clients, and we learned a ton in the process.

You know what we learned more than anything?

People who made tons of money mashing up databases of info into extremely low quality pages got a major hit. Well let's think, again, that scalable strategy worked for a while, but eventually caused a lot of (much deserved) pain to a bunch of web sites, who saw major drops in traffic. Most SEOs who embraced the hard work of adding value, saw no hits.

Let me give you a peek into what caused a consulting client to get hit by panda. We had a client that we were training, in an emerging international market, the company has hundreds of properties and millions of web pages.  It’s a HUGE task to manage this.  As we interviewed different members of their team, you could just sense in your gut that the goal was drop everything into a database and create pages from the database. This is not always a red flag, obviously large sites run this way.  The issue was that there was no focus on making these pages quality, I mean NONE.  When we would ask why does this page exist, how does it help the user…we were told not to worry about it, we pushed back, but ultimately the client didn't take the advice. 

That was almost 2 years ago, but at that time we knew that without any bloggers, no outreach, no unique content, nothing, just making pages from databases and aggressive interlinking was going to be a recipe for disaster. We knew that deep down Google & Bing don’t want this in their index. One day they got hit, why? Because they fell in love with what was easy and didn't invest in the hard work of adding value.

Scalable SEO's – Zappos is going to eat your lunch 

Are you in the clothing / retail business?  Here's an example:

Zappos is producing 60-100 videos a day according to this video! There is no way to scale that, it’s just hard work. Building studios and making the investment, and their rankings will convert better than your rankings due to that hard work!! They have proof of the conversion increases when videos are on product pages. Eventually video could become a barrier to entry / ranking signal to the search engines, and if they do, you are WAY behind. So are you waiting for that to be the case to start the investment or are you doing it NOW? 

I have been working more and more on evaluating real connections my clients have in social media, and in real life, looking at top forum posters, evaluating their top community members, or heck even looking at their LinkedIn connections as ways to build links.

Final thoughts

Many years have passed since our beloved first search engines have gotten acquired or went out of business, you know Excite, AltaVista, Lycos – even though they never became Google, they all worked towards the same goal Google and Bing do every day…reward the websites who create real businesses, good content, connect with the community and earn links with high rankings. Unlike my buddy Eric Ward, I am not 100% sold that 2012 is the year when all of this comes together, but even if it is 2013 or 2014, he is right in the fact that its coming. It’s been coming since the inception of the first search engine, I've been watching it for 13 of those years…search engines will figure it out. The question is do you want to always stay just one step ahead and have that stress or do you want to start adding value?  

Disclaimer: Every SEO (including us at SEER) has to seek out efficiencies and opportunities that scale to some extent, so I am obviously all for that. But you know when you meet someone if the core of their view of the world is more about scale and less about embracing the hard work. That is when I get worried.

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

SEOmoz Daily SEO Blog

Tags: About, Again, Algorithm, history, Never, update, Worry

Posted by randfish

I've got good news. Today marks a new Linkscape index (only 14 days after our previous index rollout) which means new data in Open Site Explorer, the Mozbar, the Web App and the Moz API. It's also more than 60% larger than our previous update in early January and shows better correlations with rankings in Google.com; I'm pretty excited.

For the past couple years, SEOmoz has focused on surfacing quality links and high quality, well-correlated-with-rankings metrics to help provide a link graph that shows off a large sample of the web's link graph. However, we've heard feedback that this isn't enough and may not be exactly what many who research links are seeking (or at least, it's not fulfilling all the functions you need). We're responding by moving, starting with today's launch, to a new, consistently larger link index.

Today's data is different from how we've done Linkscape index updates in the past. Rather than take only those pages we've crawled in the past 3-4 weeks, we're using all of the pages we've found since October 2011, replacing anything that's been more recently updated/crawled with a newer version and producing an index more like what you'd see from Google or Bing (where "fresh" content gets recrawled more frequently and static content is crawled/updated less often). This new index format is something that will let us expose a much larger section of the web ongoing, and reduces the redundancies of crawling web pages that haven't been updated in months or years.

Below are two graphs showing the last year of Linkscape updates and their respective sizes in terms of individual URLs (at top) and root domains (at bottom):

As you can see, this latest index is considerably larger than anything we've produced recently. We had some success growing URL counts over the summer, but this actually lowered our domain diversity (and hurt some correlation numbers of metrics) so we rolled back to a previous index format until now.

This means you'll see more links pointing to your sites (on average, at least) and to those of your competitors. Our metrics' correlations are slightly increased (I hope to show off more detailed data on that in a future post with help from our data scientist, Matt), which was something we worried about with a much larger index, but we believe we've managed to retain mostly quality stuff (though I would expect there'll be more "junk" in this index than usual). The oldest crawled URLs included here were seen 82 days ago, and the newest stuff is as fresh as the New Year.

Despite this mix of old + new, the percent of "fresh" material is actually quite high. You can see a histogram below (ignore the green line) showing the distribution of URLs from various timeframes going into this new index. The most recent portion, crawled in the last 2/3rds of December, represents a solid majority.

Let's take a look at the raw stats for index 49:

• 58,316,673,893 (58 billion) URLs
• 639,806,598 (639 million) Subdomains
• 135,392,083 (135 million) Root Domains
• 617,554,278,005 (617 billion) Links
• Followed vs. Nofollowed
  • 2.10% of all links found were nofollowed
  • 56.50% of nofollowed links are internal
  • 43.50% are external
• Rel Canonical – 11.79% of all pages now employ a rel=canonical tag
• The average page has 87.36 links on it
  • 73.06 internal links on average
  • 14.29 external links on average  

In addition to this good news, I have some potentially more hilarious and/or tragic stuff to share. I've made a deal with our Linkscape engineering group that if they release an index with 100+ billion URLs by March 30th (just 72 days away), I will shave/grow my facial hair to whatever style they collectively approve*. Thus, you may be seeing a Whiteboard Friday with a beardless or otherwise peculiar-looking presenter in the early Spring.

As always, feedback is welcome and appreciated on this new index. If some of the pages or links are looking funny, please let us know.

* 20th century European dictator mustaches excluded

Do you like this post? Yes No

SEOmoz Daily SEO Blog

Tags: Billion, Index, largest, Latest, Linkscape, update, URLs

Posted by randfish

If you've been following my posts on Linkscape's index, you know that we've been trying to aim for fresher, better and larger indices over the past few months, but have been finding some very tough challenges. It turns out that indexing the web, canonicalizing millions of pages and calculating a link graph with quality metrics is super-hard; who knew?

As part of those efforts, we've been working toward an experimental index that leverages a more search-engine style crawler that crawls fresher pages/sites more often and less fresh stuff less frequently. That index, however, is taking its sweet time (and we're doing a lot of babysitting and monitoring to make sure it's smooth). Our tentative plan is to have that index launched in the next 2 weeks, but we felt that since our last index was at the very end of November, a new one with fresher data was warranted. Hence, last night, we launched an interim index with the following metrics:

• 36,660,519,013 (36 billion) URLs
• 427,626,242 (427 million) Subdomains
• 128,149,029 (128 million) Root Domains
• 387,656,119,262 (387 billion) Links
• Followed vs. Nofollowed
  • 2.05% of all links found were nofollowed
  • 55.00% of nofollowed links are internal, 45.00% are external
• Rel Canonical – 10.57% of all pages now employ a rel=canonical tag
• The average page has 69.12 links on it (negligible from last index)
  • 57.76 internal links on average
  • 11.36 external links on average

This index is smalller than our last few, but the numbers look reasonably solid and the data's from the first few weeks of December, so it should be helpful to all you link builders and analyzers. Do be aware, though, that this update is likely to only last a couple weeks before we replace it with our new version, for which we have high expectations (but don't want to promise the moon just yet).

Also noteworthy – last night, when the index first launched, we experienced some wackiness with Page and Domain Authority scores. Those should have largely settled down to normalcy now, but if you see anything odd, please let us know.

Do you like this post? Yes No

SEOmoz Daily SEO Blog

Tags: Interim, January, Linkscape, update

All too often, physical distance and frantic schedules get in the way of spending time together, especially during the holidays. But you know what? They don’t have to.

With Google+ Hangouts, you can go beyond “status updates,” and connect with the people you care about using multi-person video. For free. And today we’re rolling out a number of improvements that make it even easier to say hello to your loved ones, face-to-face-to-face.

Upgrade any conversation from long-form to live
Certain posts act as kindling for face-to-face interaction. Suppose your sister gets engaged, or your roommate lands a job, or your favorite singer shares their concert schedule. You can obviously write comments back and forth, but it’s moments like these when you really want to connect in person. That’s why we’re making it easy to start a hangout from any post on Google+.

Just click “Hangout” underneath a post that you’re passionate about, and we’ll add your invitation to the comments. If others are hanging out already, you’ll see their invitation in the comments as well.

The desire to “go live” isn’t limited to your desktop computer, of course, so we’re also bringing hangouts to Google+ Messenger on mobile devices. Simply tap the new hangout icon when the time is right, and you’ll flip from text and photos to smiles and laughter.

The new Google+ app will be available in Android Market within a few days, and it’s coming soon to the App Store.

Broadcast and record what matters most
The Black Eyed Peas, the Dalai Lama and even the Muppets have all used Hangouts On Air to share their performances and peace talks with the world. On Air is still under active development, but today we’re doing three things to get us closer to general availability:

• We’re turning it on for hundreds more public figures, celebrities and other Google+ users with large followings
• We’re making Hangouts On Air completely self-service, so you can broadcast whenever you’re in the mood
• We’re integrating it with YouTube, so once you’re off the air, we’ll upload a full-length (and private) recording to your account

To get started, just look for the new “broadcast and record” option after starting a hangout. If you don’t see it yet, then don’t worry: we’re working hard to give everyone the chance to go On Air.

In the meantime, you can still watch or join On Air broadcasts directly from the stream. Just look for the red banner while reading your posts or checking your hair, and you’ll know you’re in the right place.

Dial-in friends and family, from all over the world
Not everyone has a webcam or a front-facing mobile camera, but that shouldn’t stop them from spending time with the people they’re close to. Fortunately, nearly everyone has a telephone, and today we’re making it possible to dial-in anyone, from almost any country, directly into your hangout. Calls to the U.S. and Canada are free, and international calling rates are super, super low.

Never miss a chance to connect
When friends and family invite you to hang out on Google+, we want to make sure the opportunity doesn’t pass you by. So we’re making active hangouts more accessible in two important ways:

• On the right-hand side of the stream, you’ll see up to three live hangouts that you can join
• Whether you’re at your desk or on the go, we’ll ring your device and notify you when you’re invited to a hangout

Whether it’s sharing baby news, or your niece’s new dance moves, Google+ users continue to use hangouts to build intimate onscreen experiences. We hope these two improvements help you discover more of them, more often.

Have some fun, add some antlers
Back in NMovember, we helped the Google+ community add millions of virtual moustaches to their faces. Now that it’s December we’re introducing a pair of antlers, so by all means, unleash your inner reindeer! Looking ahead, developers will be able to create their own effects using the updated Hangouts API, so stay tuned for lots more fun.

By bringing people together, face-to-face-to-face, we’re hoping to make the world a bit cozier, and lift people’s spirits a bit higher. So give hangouts a try this holiday season, and let us know what you think.

Posted by Vic Gundotra, Senior Vice President

The Official Google Blog

Tags: Beyond, Going, Google, Hangouts, Status, update

Posted by randfish

Hey gang – it's that magical time again when Linkscape's web index has updated with brand new data (for the second time this month). Open Site Explorer, the Mozbar and the PRO Web App all have new links and scores to check out. This index also features the updated Page Authority and Domain Authority models covered by Matt last week on the blog.

Here's the current index's metrics:

• 38,295,116,929 (38 billion) URLs
• 466,742,600 (466 million) Subdomains
• 125,007,049 (125 million) Root Domains
• 387,379,700,299 (387 billion) Links
• Followed vs. Nofollowed
  • 2.03% of all links found were nofollowed
  • 55.57% of nofollowed links are internal, 44.43% are external
• Rel Canonical – 10.34% of all pages now employ a rel=canonical tag
• The average page has 70.61 links on it (down 6.67 from last index; we're likely biasing to a different set of webpages with the broader vs. deeper focus of this release)
  • 59.02 internal links on average
  • 11.59 external links on average

As you can see, we're crawling a LOT more root domains – we expect to have data for an extremely high percentage of all the domains that you might find active on the web. However, because of this broader crawl, we're not reaching as deeply into some large domains (some of that is us weeding out crap, including many more millions of binary files, error-producing webpages and other web "junk"). You can see below a chart of the root domains we've crawled in the last 6 months vs. the total URLs in each index.

We work toward a few key metrics to judge our progress on the index:

• Correlations with Google rankings (not only of PA/DA, but of link counts, linking root domains, mozRank, etc)
• Percent of successful API requests (meaning a request for link data on a URL from any source that we had link data for)
• Raw size and freshness (total # of root domains and URLs in the index, though, as Danny Sullivan has pointed out, this may not be a great metric on which to judge a web corpus)

We've gotten better with most of these recently – PA/DA have better correlations, more of your requests (via Open Site Explorer, the Mozbar or any third-party application) now have link data, and we're slowly improving freshness (this index was actually completed last week, but didn't launch due to the Thanksgiving holiday). However, we are not improving as much on raw index size (root domains, yes, which we've seen correlate with other metrics, but raw URL count, no). This will continue to be a focus for us in the months to come, and we're still targeting 100 billion+ URLs as a goal (though we're not willing to sacrifice quality, accuracy or freshness to get there).

As always, if you've got feedback on the new scores, on the link data or anything related to the index, please do let us know. We love to hear from you!

Do you like this post? Yes No

SEOmoz Daily SEO Blog

Tags: Broadest, Index, Live, November, PA/DA, Scores, update

A new Google algorithm update designed to pick and choose which queries demand fresher results impacts about 35 percent of searches. Where their Caffeine architecture allows Google to quickly crawl the mass and complex contents of the Web, th…

Search Engine Watch

Tags: Affecting, Algorithm, Fresh, Gets, Google, Searches, update

Posted by caseyhen

Yesterday, Google announced that they released a new update that impacts roughly 35 percent of searches and can better determine when to give you more up-to-date relevant results. What does that mean for you as a search marketer? Rand, with special guest Mike King (@iPullRank) dive into what this mean for you and your clients. Let us know your thoughts on this most recent update in the comments below!

p.s. from Rand: I’d also recommend reading this excellent post from Justin Briggs on Methods Google May Use to Evaluate Freshness.

Video Transcription

Rand: Howdy, SEOmoz fans. Welcome to a special edition of Whiteboard Friday. This week I am joined by none other than Mike King. Mike, so great to have you.

Mike: What’s going on, Rand? Thanks for having me.

Rand: Mike is out of Publicis in New York. He spoke recently at the Search Love Conference and spoke here at the Seattle Interactive Conference.

Mike: Absolutely.

Rand: Getting big on the scene. And Mike, we’ve got some supposedly big changes from Google coming out today.

Mike: We had the Google Freshness update today.

Rand: So Google announced it, right? So Amit Singhal, head of Search Quality at Google, writes on the blog and he says a few things, and we’ve got some questions about this. He says, "Google’s new Freshness Update affects 35 percent of queries. It prioritizes recent and timely results, and it’s based off their Caffeine infrastructure."

Mike: So far what we’re seeing is it’s 35 percent of queries, but I think people are expecting that to mean 35 percent of keywords. That’s not what’s happening. So we’re seeing it on a lot of head terms. For example, here’s a SERP that we saw for football. What we’re seeing is really recent posts being annotated to the SERPs, so they’re having 8 hours ago, 3 hours ago, 18 minutes ago.

We’re also seeing that for basketball, Microsoft Courier, Wall Street, and Top Chef Texas. You can google those right now and see these in the SERPs.

Rand: When they say affects 35 percent of search results, and we’re seeing, like, boy, it feels a lot more subtle than Panda. A lot of SEOs are like, "Boy, 35 percent of queries. You said Panda only affected 11 or 12 percent." Something feels disconnected. Talk about the difference between affecting keywords versus affecting query volume.

Mike: Right. When they’re saying 35 percent of queries, these are words that people are actually searching for. It’s not necessarily just every keyword in the keyword universe.

Rand: Gotcha.

Mike: So it could be a much smaller set of keywords than we’re talking about here.

Rand: If 10 percent of people in the world google "Kim Kardashian wedding," tragically – they really shouldn’t, but they do – then that could be a huge part of what they’re saying is affected here.

Mike: Absolutely. That’s what we’re seeing, a lot of celebrity keywords that are being affected. But that kind of makes sense because those are inherently QDF keywords.

Rand: Right. Query deserves freshness. One of the things that we noticed is it seems to help lots of date-specific content, not just hyper- new, meaning some of these results are 8 hours ago, 3 hours ago, but we’re also seeing a lot more stuff that’s . . .

Mike: It’s like two days ago or seven days ago, but it’s all date-specific, like you’re saying.

Rand: Yeah. Then there are some of these new annotation types of results. So this, fundamentally, looks different to us. It’s not site links.

Mike: Absolutely. They’re direct links to individual articles rather than site links.

Rand: And we think these are RSS-based. Is that right?

Mike: Absolutely. So we did a few, like, poking around a few different feeds and things, and we saw that they did match up almost directly with the RSS feeds.

Rand: So if you’re trying to illustrate specific content in your fresh links, which Google is now providing you an opportunity to do, RSS seems to make a ton of sense.

Mike: Definitely. For example, they have the last mod . . . what’s the word?

Rand: Oh, the last modified date? You mean the stamp?

Mike: Right, exactly. So that time stamp seems to be something that’s affecting that. So keep those up-to-date, and your XML sitemaps could definitely help with this.

Rand: Right. We were talking about the Top Chef Texas query. And when you look at the Top Chef Texas page today, it’s basically Bravo TV, and then every result in there is from the last 12 hours.

Mike: Absolutely.

Rand: Meaning, Mike, you and I write a great blog post about Top Chef Texas last week, and it’s ranking well. Today, forget about it.

Mike: Yeah, you really have to stay on top of it. So if you’re writing content about Top Chef Texas, make sure that you continue to have content on it. That’s like a TV show that happens regularly. If you’re going to do that, you need to write about every episode to stay on top of it.

Rand: Yeah. And this is kind of a big change for Google.

Mike: Definitely.

Rand: So that could be part of that 35 percent that it’s affecting.

Mike: Absolutely.

Rand: I’m going to make you come this way with me. We’ve got some takeaways for marketers here. One of the ones, tell me about this, "Watching Your Important SERPs for Signs." How do I check this? How do I check whether I’m going to be affected, and what search results should I be watching? If we’re talking about most important SERPs, we’ve got to thinking about things that drive the most conversions, the thing that drives the most traffic, and the thing that drives the most engagement. And then for fresh and time saved content, basically if you’re seeing . . .

Mike: Blog posts.

Rand: Yeah. If you’re seeing the eight hours ago, you’re seeing the days, you’ve got to do that. I mean, you and I are both huge believers in this, right?

Mike: Absolutely. Content marketing, if you’re not doing it, like it’s says, you’re crazy.

Rand: You crazy. I wrote that there.

Mike: He did. He did.

Rand: And then finally, we’ve got to watch whether and how Google is grabbing this time stamped content. We have some questions about the RSS feeds are being pulled in here. It doesn’t always seem to be just the most recent items.

Mike: It doesn’t seem to be like a one-to-one thing, so it’s hard to say. Maybe they’re looking at shared or how much is being pushed on social.

Rand: Google+.

Mike: Exactly, Google+, how many people are pressing that magic +1 button. So it’s hard to say, but I would venture to guess that they’re thinking about social, because if it’s the most relevant, it makes more sense that it’s shared the most.

Rand: And if it’s coming from RSS, they’ve got have all that data about who’s clicking on those RSS feeds.

Mike: Absolutely. Like Will Reynolds was saying, they own FeedBurner, so they have all this data. Are people reading this stuff?

Rand: Sure.

Mike: So this may be that final application of it.

Rand: Fascinating. All right. So supposedly there’s a huge change, but we’re not feeling it nearly as dramatically as Panda.

Mike: Not at all.

Rand: So we’ll see how this rolls out. I certainly look forward to comments and feedback from you guys. Mike, thank you very much for joining me. Appreciate it.

Mike: Thank you for having me.

Rand: Cheers, gang.

Video transcription by Speechpad.com

Do you like this post? Yes No

SEOmoz Daily SEO Blog

Tags: &quotFreshness&quot, Friday, Google's, update, Whiteboard

On Tuesday I reported how the Google Panda 2.5.2 tweak was more than just a “minor” update.
There were just too many complaints in the forums for it to appear to be really that “minor.”
So I decided to run a poll, where we have well over 400 responses. The responses were overwhelming against the notion that this was a minor update or tweak to the Panda algorithm.

Search Engine Roundtable

Tags: Google's, Minor, Panda, Says, Significant, update, Very

Google has confirmed that the latest iteration of its Panda algorithm update is live. Based on our tracking of the algorithm changes, this is Panda Update 2.5. Google declined to share any specifics about what types of sites, pages or content this update targeted, instead only sharing the…



Please visit Search Engine Land for the full article.

Search Engine Land: News & Info About SEO, PPC, SEM, Search Engines & Search Marketing

Tags: Arrived, Confirmed, Google, Panda, This, update, week

Posted by Bryce Howard

Howdy folks! I wish I was writing you with better news, but in the spirit of TAGFEE, we want you to be as informed as possible about your PRO membership: 

Due to a major PRO web crawler service outage that occurred on Friday evening, crawler-related PRO features (link analysis and crawl diagnostics) are currently disabled. However, rankings, on-page optimization, and all tools except Crawl Test are functional.

So what the *bleep* happened!?

Amazon turned the lights out on us. Well, not exactly—I’ll explain. We host a number of our web applications from Amazon Web Services (AWS). For many of these hosts we pay a fixed rate per hour, however AWS offers an alternative billing model called spot instance pricing. Spot instance pricing is a method for purchasing excess computing power from AWS at a respectable discount. Everybody wins, we get a great price for the hundreds of computers we use daily while AWS is able to sell a resource that’s otherwise just sitting around idle.

But the use of spot instance pricing comes at a risk: the computers hosting your services are only allocated to you as long as there is still excess capacity and that no one else is willing to bid more for those hosts than you are. If someone comes along offering to pay more, then AWS may revoke your hosts without any warning, leaving you to rebuild your services from scratch. This is not so bad if you can ensure you have enough computers left to still service requests… and therein lays the problem. 

Our Mistake

The contract of spot instance pricing is quite clear: your servers may be arbitrarily taken from you, so you must be strategic about its usage. We unfortunately did not apply good strategy to our PRO web crawler configuration. Almost all of our service hosts were spot instances allocated with a dangerously low bid price (e.g. $ 2.00/hour), and they were all clustered within the same AWS availability zone (more on this later).

So we put ourselves at risk with a low bid price, excessive use of spot instance pricing, and a poor distribution of hosts across AWS availability zones. We bet that there’d be little to no chance that AWS would reclaim our spot instance hosts but we bet very wrong. At approximately 6 PM PST, AWS terminated approximately 50% of our active spot instance hosts in the PRO crawler service cluster. Around this same time, the going spot instance price shot up to $ 2, our maximum bid price, which triggered this culling of our service hosts.

Losing half our hosts wasn’t entirely catastrophic, however bad. In fact, it was a salvageable situation but then it got much worse. At 9PM PST we lost all of our service hosts that were spot instances (> 90%). The going spot instance price had jumped to $ 2.51/hour at this time, and given most of our hosts were bid at the price of $ 2/hour we effectively forfeited all rights to our previous claims. Our service wasn’t broken; it was just plain gone!

This pretty graph from AWS accurately documents the spot instance pricing timeline for the day in question:

How We Could’ve Prevented This

Three practices that could have prevented this from occurring:

1) Use a spot instance price that is commensurate with the value of the service.

If a host was very critical to service functionality, we should’ve bet a much higher price than the $ 2/hour. Using the spot instance pricing chart as a guide we should have at least used a bid of $ 3/hour or more to ensure better chance of avoiding host reclamation by AWS.

Could we have predicted this optimal bid price? Likely not. Regardless, we should’ve bid what we thought the continued functioning of our service was worth. I think it’s easy to appreciate we now find that value much higher than the $ 2/hour we’d originally bid.

2) Distribute hosts across multiple availability zones.

The initial increase in spot instance price occurred in one availability zone (us-east-1c), with the secondary increases in us-east-1c and us-east-1d. Had we spread our bets across multiple availability zones, we could have weathered this price volatility with at least half of our service hosts intact, even at the bid price of $ 2/hour.

Although we were aware that prices could vary by availability zone, we did not use this to hedge our bets more effectively.

3) Use a mix of on-demand and spot instance pricing.

On-demand priced hosts use a different pricing strategy where you agree to pay a fixed amount per hour to AWS but in return you get certain guarantees about your host claim, most notably it won’t be arbitrarily taken from you due to demand. Had we diversified our portfolio between on-demand and spot instance pricing we could’ve ensured at least minimal functionality of our service in the worst case while enjoying some good amount of cost savings in the best case.

As with any critical investment you have to be strategic about minimizing your downside; we will do this moving forward.

So where are things?

To be frank, we are absolutely mortified that we’ve had to disable such an indispensable product feature as crawl diagnostics, especially when this service outage was otherwise avoidable. We are literally working day and night to re-enable the PRO app crawler service. Currently, we are rebuilding the API servers, the underlying NoSQL data store (Cassandra), and the various processing and crawling hosts. We are being very careful as we do this to avoid the previous mistakes, being strategic about diversifying pricing type (on-demand vs. spot-instance), distributing across availability zones and using a very competitive spot-instance bidding price.

Most of the aforementioned service components are pretty easy to restore, but we have one unfortunate problem that will somewhat delay full restoration of the service: the terabytes of data generated by the hundreds of thousands of crawls we’ve executed over the last nine months. We must load this data from our backups (securely stored in AWS S3) into our NoSQL data store, something that by no means can be done quickly.

Being perfectly transparent, this is an operation that could take the full duration of a week. We certainly don’t want to make anyone wait a full week just to see data that’s already a week out of date, so we plan to be a bit more clever with this service restoration, choosing the most optimal path to populate our database while also ensuring we preserve our weekly crawl cycle. Do we have all the solutions in place to achieve these goals? Not immediately, but we are making great progress and I’m very confident we will have more optimistic projections about service restoration in the next several days.

Ok, so how exactly does this affect me again?

As a PRO member you can still:

• Create new campaigns
• Check your rankings
• Manage keywords
• Check your on-page SEO
• Run reports
• Check your backlinks & traffic data
• Use Open Site Explorer
• Watch webinars
• Ask & answer questions in PRO Q&A

For the next week you won’t be able to access:

• Crawl diagnostics for any of your campaigns

Also as a reminder, none of the data is lost, we just need time to rebuild so we can access it.

In the meantime, rest assured that we are doing everything we can to get your PRO functionality back up and running like it’s meant to be. We realize that many of you rely on this data to optimize your company’s and clients’ sites, and want to return service ASAP so you can continue to do what you do so well. Thank you for hanging in there with us as we learn from our mistakes.

Do you like this post? Yes No

SEOmoz Daily SEO Blog

Tags: Crawl, Doing, Outage, update, were