Google is one of the most widely misunderstood success stories of our time. Many of us equate Google with “Big Data,” that is, amassing huge quantities of data and then finding useful statistical patterns. But is that how it succeeded?
In Life after Google: The Fall of Big Data and the Rise of the Blockchain Economy, George Gilder criticizes Google primarily on two fronts: First, it is a “walled garden,” a great platform, but inherently isolated and closed. That is a point worth exploring, but not the focus here. The second point, the one I want to touch on, is that Big Data’s day has come and gone. Because Google is a Big Data company, its brightest days are behind it.
First, Google did not succeed by being a Big Data company. It is unclear if the company even knows this about itself: It has had innumerable failures, and its failures are almost all Big Data-oriented. Where Google has succeeded is by finding, marking, and harnessing human intention (intentionality), not data.
To understand how this works, think back to the Google’s early days (founded 1998). It actually had the smallest dataset of the competitors in the search engine field. Other companies could find more words faster. Other companies had better document searches. So what made Google a better search engine in the long run?
Google’s original success came from harnessing the linking structure of the web. But where do these links come from? From nature? No. Links unearth a sea of human intentionality lurking in the pages.
While people can and do say almost anything about themselves on a web page, they rarely refer users elsewhere unless they really want those users to go and check out the other page. After all, they are asking you, the user, to go and pay attention to something other than themselves. Therefore, the links provide a way to identify the greatest level of human intentionality. The famed Google PageRank algorithm is Google’s first method of harnessing intentionality.
It turned out, Google didn’t need the biggest database or the fanciest statistical algorithms to win. Google won because it recognized an easy way to identify and magnify the intentionality expressed in web pages.
Google’s second big achievement was their Adwords system (now just called Google Ads). Banner ads predated Adwords by six years (history here). Doubleclick was one of the big players in the banner ad space prior to Google. GoTo.com (eventually acquired by Yahoo) offered paid search results. This approach is the essence of Big Data applied to ads.
Google Adwords originally began with this approach. However, Adwords overtook its competitors by harnessing user intentionality. While other services would show users an ad as long as it was paid for, Google Adwords would show only ads that users were clicking on. In other words, they were tracking and responding to user intentionality towards the ads. The ones which users interacted with were kept in place and the other ones were dropped.
Interestingly, Google continues to improve its searches with the same method, “intentionality harnessing.” It detects whether or not you click on a link that it has served you. Therefore, it knows if a result was relevant to your search. It can sometimes even detect how long you spend on the search result. Again, that’s Google harnessing your intentionality to improve its results.
As we’ve noted before, one of the best ways for machines to learn is to utilize the human mind for input and training. In the same way, while Google does have a large pile of data and sophisticated statistical methods, its true secret sauce is not the size of that pile but the ability to detect, quantify, and utilize the intentionality of users. It has harnessed one of the powers of the human mind.
Jonathan Bartlett is the Research and Education Director of the Blyth Institute.
“Artificial” artificial intelligence: What happens when AI needs a human I?
Note: For more on the concept of intentionality, an immaterial power of the human mind connected with free will, see Michael Egnor, Do either machines—or brains— really learn? and Neurosurgeon outlines why machines can’t think.