Posts Tagged ‘Algorithm’

Google Goes Live with Fresh Algorithm Change

November 7th, 2011

Once again, the developers at Google have opened their toolboxes and made a few fixes to the world’s most reliable search engine.

The changes to the Google ranking algorithm were made live yesterday morning. Did you hear it? When the news broke, there was an almost-perceptible hush as silence descended on cubicles in offices across the globe – followed by a calamitous rattle as millions of cans of RedBull clattered to the desktops. Propelled to action by the notion of a change in the almighty algorithm, SEOs and Web Marketers across the planet dropped whatever they were doing and began furiously typing in search strings to see the search results.

The results were fast and fresh. So, why the alarm? Well, outside of the notion that Google had tweaked their algorithm to index, crawl and return results faster, developers from HQ in Mountain View, California also stated that the improvements to the ranking algorithm impacted roughly 35% of all searches. Again, that’s 35%, people – a larger percentage than the Panda update which only impacted 12% of conducted searches. The concern, as you may imagine, was that a whole bunch of search results just lost their expected placement. This is a good thing.   

google-fresh-algorithm

The present change to the ranking algorithm supplements and supports the Caffeine web indexing update that was completed by Google in June of last year. The Caffeine indexing system was built in response to the changes in available content on the web. The developers at Google (and just about everyone else who spends a little time on the web each day) know that content on the web is both thriving and growing at an alarming rate. More content means more complexity. More content also means greater expectations from content producers and searchers alike. Google built Caffeine to index pages on an enormous scale, adding new information to the database at a rate of hundreds of thousands of gigabytes each day. Then they added in the freshness factor.

The primary term that Google has created over this new algorithm change is freshness. By this term, Google developers are expecting the algorithm to determine when to give searchers more up-to-date, relevant results for varying degrees of information freshness. Simply put: given the absurd rate at which information moves and changes, different search types must have different freshness requirements – if the search results intend to give the searcher up-to-the-minute information.

Examples:

  1. Under the new update, a search for information about a recent news event, such as the “Occupy Wall Street” protest, may return pages with information only a few minutes old.
  2. A search for a regularly-occurring event, such as the World Series, would prompt search results for the most-recent World Series (Cardinals over the Rangers in Game 7) and not some historical game.
  3. Those searches for information that receives frequent updates, such as consumer product reviews on laptop computers, assume that the searcher probably wants the most up-to-date information on the brand and not something written three years ago for a product that is no longer manufactured.
  4. On the other hand, we did find a 2006 Matt Cutts blog post ranking #1 for “URL canonicalization“. This content ranks above 2009 content from SEOmoz, and 2010 content from Vertical Measures.

I think we can conclude that Google has favored freshness for sometime, but not discounted all old content completely. Links, comments, social signals, and general URL trust/freshness are other indicators that your content should still rank. This will keep some old content rankings strongly, as we found in example four above.
Was it a good update to the algorithm? Did the world as we know it change dramatically?

Somewhat, yes.

The freshness factor has already been shown to impact one out of every three searches (hence the 35%). Some experts have suggested that the freshness factor may open search results to new problems with relevancy, search engine poisoning and junk content. In order for the changes to work as the developers wish, Google may have to leverage a few other search ranking factors as future tweaks to the freshness update.

SPAM Farms get the Plow from Google

January 28th, 2011

“You ain’t gonna work on Maggie’s spam farm no more.”

Fed up with sites that produce “shallow or low-quality content,” Google has finally released a new spam detection classifier to help prevent on-page spam content from ranking in the Google search results pages. And although the news was met with cheers and tears (given some collateral damage), we all knew this was coming.

Matt Cutts, head of the Webspam team at Google, earlier hinted at this spam-catching action last November – during his talk at the PubCon search anGoogle-matt-cutts-spam-costumed social media conference in Las Vegas. A long-time opponent of spam (in all its unsavory forms), Cutts vilified sites that scrape and copy original content from other pages as well as those sites that offer negligible levels of original content. In his talk, Cutts promised that 2011 would be the year that Google invested in fresh spam-catching efforts and that they would also take a hard look into identifying hacked sites that routinely push poor content into the SERPs.

So, on January 21st, the Webspam team at Google, made good on their earlier threat to enact new techniques to fight content spam and altered the algorithm to drive down spam levels with the new classifier.

So, does it work? Maybe.

Some early detractors believe that this is job too detailed to give over to an automated algorithm. Others think that Google is having a difficult time with even basic link schemes. In the end, Google has the last word. Eliminating content spam sounds like a great idea. Only time will tell if it’s truly effective and beneficial to Google’s searchers. Clearly, Cutts thinks it’s time for a change. And his views are often driven by feedback he receives from SEOs and search users alike.

“People are asking for even stronger action on content farms and sites that consist primarily of spammy or low-quality content,” said Cutts on his recent blog post. “We take pride in Google search and strive to make each and every search perfect. The fact is that we’re not perfect, and combined with users’ skyrocketing expectations of Google, these imperfections get magnified in perception. However, we can and should do better.”

When it comes to sorting out webspam or content spam, Cutts explained that the new classifier is more adept at identifying spammy content and spammy words on derivative, self-promoting web pages and blog sites. Google also improved their ability root out hacked sites that are pushing spammy content.

Cutts himself defined the content farms as those with “shallow or low-quality” content and has promised to evaluate forthcoming changes that may help drive spam levels even lower.

For those of us in the content and search marketing industry, it’s common knowledge that original, relevant information will always increase interest, traffic and profits. The new spam detection classifier may be the first of numerous changes from Google that will benefit us all. But, as stated before, only time will tell.

Yes, our opening line was a tip of the hat to “Maggie’s Farm” by Bob Dylan.

Update 2-9-2011:

Google Webmaster Tools has a new HTML message that talks specifically about monitoring for duplicate content. Here is a screen shot:

Google-Webmaster-tools-dup-content-message

Update: This Topic Was Well Covered at SMX West 2011.

Google Instant – What’s All The Hype About?

October 15th, 2010

By Q3 of 2010, Google’s main search site had passed the one-billion-per-week user mark. That number makes Google one of the most-used services – of any kind – in the world. It’s a significant achievement and one that had the Google Development Team talking about scalability and processing long, long ago.

To commemorate this milestone, Google held a media conference at the Museum of Modern Art in San Francisco on September 8th and introduced Google Instant.

For those unaware, Google Instant is a new search enhancement that serves to improve the quality of search functions and the overall user experience by showing search results as you type – updating search results and delivering ads with each new keystroke. It’s a predictive search with real-time visualization and results. The impetus for this enhancement comes from a need to reduce the time it takes to log a search query at Google – thereby saving valuable processing time.

In the research and development phase, Google’s usability study showed that most people type slowly but scan information for content and results extremely quickly. With this in mind, the Google Development Team expected that most Google users would be able to scan for results they need – while still typing. Google Instant was born.

Clever as this new instantaneous search enhancement may be, Google has some in the search world a bit nervous – especially those in the SEO or PPC professions. Throughout all of the glorious fanfare and inevitable pushback, Google maintains that nothing has changed with search result ranking.

There’s no reason to be nervous. Google Instant is certainly not the first enhancement to the world-renown search engine. So far, in 2010, Google has already rolled out over 500 changes to ranking and UI. And – until all the speculative chatter can be thoroughly tested and proven either true or false, much of what we’re hearing is just chatter. Not only that, but if you don’t like it, Google made it possible for users to turn it off.

Here is a short list of what some are talking about:

1. Google is rendering the paid and universal search listings as high as possible, but may be pushing the SEO results “below the fold.” This, of course, may mean that top ranking is of greater importance than ever.

2. With the prominent placement of the PPC results in the Google Instant page rendering, there may be some impact on PPC performance. Some expect that the nature of the page rendering will drive CPCs up and CTRs down.

3. What about quality score? Google has stated that Google Instant does not change the way they determine the relevance or quality of your ads. Advertisement performance will be judged as it has been in the past.

4. Monitoring key performance indicators related to SEO and SEM should remain the same. Google Instant search is supported by Chrome v.5 and v.6, Firefox v.3, Safari v.5 for Mac and Internet Explorer v.8. However, Google Instant search is not supported in Internet Explorer versions below 8 or Firefox versions less below 3 – so you may wish to filter your metrics accordingly.

Above everything that will be said about Google Instant in the months to come, it should be mentioned that Google Instant is important enhancement for search users and advertisers alike. And much of the initial negative information about Google Instant is, at this time, mere speculation.