Previously, we talked a bit about how the best search engines are like the best newspapers: focused on building trust with their audience. This deep trust helps search engines build their audience, and increases the value of their text ads.
But the comparison between the newspapers and search industries goes beyond building trust. Both industries actively manage what they call their “news hole”: the percentage of material in the newspaper that is editorial content, not paid advertising.
Newspapers think about their news hole every single day: if it gets too low, then subscribers will feel overwhelmed by ads. If it gets too high, than they’re not making enough money. Warren Buffett got this right away – here he is (back in 1983!) explaining how the Buffalo News thinks about their news hole:
The [Buffalo] News lives up to its name – it delivers a very unusual amount of news. During 1983, our “news hole” (editorial material – not ads) amounted to 50% of the newspaper’s content (excluding preprinted inserts). Among papers that dominate their markets and that are of comparable or larger size, we know of only one whose news hole percentage exceeds that of the News.
Google very actively manages their news hole: the percentage of links on the page that are “organic search” versus “paid text ads”. When I search for “Warren Buffett”, I’m not surprised to see that Google’s news hole is exactly where Warren would like to see it: 50%. And even more importantly than that number is where the ads are: consistently separated from the organic search, and isolated on the right side of the page.
Ask.com on the other hand, seems to have done a poor job at first at separating its paid ads from its search results. I’m not an expert on the history of Ask Jeeves, so I’ll have to rely on this this June 2004 forum post by SEO guru “Black Knight”:
Ask [Jeeves] used to provide some very different results for a while (back in around 2000) which while not relevant enough to most questions to make it a primary engine choice, did make it one of my backup engines for further research. It suffered the same fate that so many dot-com boomers did – it was wrecked by over-eagerness to make money. In no time, Ask was filled with nothing but PPC [Pay Per Click] listings, and effectively killed off for anything but shopping related searches.
Filling the Newshole
There are two ways to fill a newshole: with Content, or with Algorithms.
CONTENT
Content is the most traditional way to fill a newshole.
Newspapers get their content the old fashioned way: they hire reporters and editors to write articles. All those salaries can add up, so newspapers have come up with cost-saving measures to buy content ala carte. Comic strips, advice columns, crosswords can all be bought from syndicators like United Media. But probably the biggest syndicator of content is the Associated Press or Reuters. Most newspapers these days fill most of their newshole with cheap AP wire copy.
But professional editorial content isn’t the only form of Content out there. As Journalists become an increasingly rare breed, UGC (User Generated Content) is stepping up to fill the content vacuum. RSS Aggregators are an interesting example of this, as sites like Bloglines fill their newshole almost entirely with weblog posts. Gmail is another example, as it fills its newshole with user generated emails.
It’s a whole new world…
ALGORITHMS
This is where things start to get crazy. As it was mentioned earlier , Search Engines fill their newsholes with organic search results. Why is that so crazy? Because Google didn’t write a single word of that content! They get the content for FREE from websites like the New York Times and the Washington Post. The value of Google is in the search algorithm itself: by calculating which piece of content you’re most interested in, Google earns its place in its users’ lives (and gets a hugely lucrative news hole in the process).
It’s a big disingenous for me to say that the the content is free. The content may be free, but that doesn’t mean that Google’s newshole itself is free. The cost of researching Google’s search algorithms and building outs its enormous data centers is in the hundreds of millions of dollars (if not billions). But the nice thing here is that the cost of Google’s Newshole is largely a fixed cost. The variable cost of generating the search and textads itself is neglible relative to the revenue involved. No wonder Google’s margins are looking so good these days…
One last example of an Algorithmic News hole: Bloomberg! Here’s a quote from Michael Bloomberg’s memoirs, where he describes his Bloomberg News division:
For some stories, we don’t use human writers or editors at all. When we describe a market’s value at a given instant (as opposed to why it had moved there), only speed and accuracy matter – and they are not most people’s strengths. …
We’ve programmed our Bloomberg computers to “write” periodically a series of stories informing readers of the current state of the market. … A sample result: “The Dow Jones Industrial Average was 1.09% lower at 3:01 p.m. Eastern time, down 62.14 at 5650.24. The stocks that contributed most to the average’s fall were Procter & Gamble Co., which lowered the Dow by 5.55 points; General Electric Co., 5.18 points; and the Walt Disney Co., 4.07 points”
Bloomberg’s automated stories are as algorithimic as Google’s Search newshole! There are lots of other examples of algorithimically generated content (e.g. Google News, Google Alerts, Amazon’s section pages), that show how algorithimic newsholes are becoming increasingly common.
So you can fill your newshole however you like… with human-generated Content, or machine-generated Algorithms. Increasingly, it seems like the future belongs the cheapest of each: User Generated Content and Algorithmic Content.
In other words: blogs and search. Things are starting to get very interesting…
This article was written by John Hiler