Blind Five Year Old

There’s still a lot of speculation and conjecture about Google Caffeine. More than a few have analyzed and theorized based on the preview Google provided. Some have even given tips on optimizing for Caffeine. Don’t believe the hype.

The Google Caffeine Myth

Google Caffeine is not an algorithm change. That’s not to say that the results won’t change here and there, but for the most part the actual scoring is unchanged. There are no special techniques or changes needed to address Google Caffeine.

What is Google Caffeine?

Google Caffeine is a re-architect of the entire indexing system for performance. In short, Caffeine:

• allows Google to crawl and index more pages
• allows Google to crawl and index pages faster

The result? The indexing data fed into individual signals is more robust and updated more frequently. This might produce some slight changes to results, but only because of the change in data, not in the change in scoring.

In Google terminology, Google Caffeine increases the rate of data refreshes. I recommend you read and listen to what Matt Cutts has to say on the topic. Vanessa Fox was also a voice of reason with her Search Engine Land Google Caffeine review.

Now, in re-architecting the indexing system, not all data structures are 100% compatible. So, some tiny changes have likely been made, but they’re not meant to be algorithmic changes.

Why Ask For Caffeine Feedback?

Google provided the SEO community with access to Caffeine results and asked for feedback. This request was looked upon by many as an indication that this was an algorithm change.

In fact, this request was a type of distributed QA. The goal of the new indexing system is to increase performance but not appreciably change the algorithmic results. Letting a bunch of hyperactive SEOs into the sandbox helped them to identify any flaws to their re-architected code.

Why Re-Architect The Indexing  System?

Most seem to believe the impetus for Caffeine was around real-time search. I’m sure this was factor, but I doubt it was the primary reason. Google could have pulled off real-time search without Caffeine.

The indexing code in use was written ages ago in Internet-time. Since that time, the Internet has grown and changed dramatically. Algorithmic changes and new signals could only go so far in ensuring quality results. If the data set they were using was incomplete or ‘old’, no amount of signal tinkering would have the desired impact. It’s a simple GIGO problem.

Google needed more data to better inform the algorithm. They need to see more of the Internet. More links. More page changes. Any of you who have ego based Google Alert feeds would have noticed a substantial increase in activity over the last month. It’s not that you’re getting more popular (sorry), Google’s just indexing more and doing so faster than ever before.

Caffeine Is The Beginning

The real question should be what will Google do once Caffeine is fully implemented. Once Google gets all of this data, new patterns will emerge and algorithm changes are certain to follow.

Google wants to see long clicks not short clicks. That’s the new SEO terminology coming from Google.

What’s a Long Click?

A long click occurs when a user performs a search, clicks through on a result and remains on that site for a long time. They don’t come back to the result set immediately to click on another result or to refine their query. In general, long clicks are a proxy for satisfaction and success.

What’s a Short Click?

A short click is the opposite of a long click. (No duh!) A short click occurs when a user performs a search, clicks through on a result and quickly comes back to the result set to click on an alternate result. In general, short clicks are a proxy for dissatisfaction and failure.

How is Google Tracking Click Length?

No, it’s not through Google Analytics. They’re not peeking at bounce rates. Instead Google is measuring pogosticking activity by leveraging their current tracking mechanisms. Remember, Google already tracks the user, the search and the result clicked. All Google needed to do was to accurately model the time dimension.

Why are Long Clicks Important?

Long clicks are important to Google because it gives them a way to measure the satisfaction of the result based on downstream behavior. Sure, a result might get a lot of clicks but did it actually satisfy the query?

Is it a success if 100 people click but 98 go back within 10 seconds? What if those 98 people all clicked on an alternate result?

Google knows that the search algorithm still isn’t that smart. It routinely makes mistakes and can often be led down the wrong path by aggressive search engine optimization. Long clicks provide a feedback mechanism, a type of human quality assurance that is lacking in the algorithm.

Long clicks are important to you because they will may help increase your SERP rank.

The chatter from Google makes me believe that it is part of the algorithm. How much it is weighted now and in the future remains to be seen. One way or the other Google is saying that longer is better.