Nieman Foundation at Harvard
Postcards and laundromat visits: The Texas Tribune audience team experiments with IRL distribution
ABOUT                    SUBSCRIBE
March 9, 2011, 1 p.m.

Richard J. Tofel: Someday, the sun will set on SEO — and the business of news will be better for it

Editor’s Note: Richard J. Tofel is general manager at ProPublica, a Wall Street Journal veteran, and author of a number of books, most recently Eight Weeks in Washington, 1861: Abraham Lincoln and the Hazards of Transition. Here he looks toward a future when search engine optimization has been rendered obsolete by advancing technology — and the implications for news.

The first time I saw the Google guys in action, one of them — I believe it was Larry Page — stunned the small crowd. It was long before the IPO, when Google was the Next New Thing, the search engine that the cool kids in the class, or the office, were knowingly mentioning to the rest of us. The interviewer was fawning over the young upstarts when Page said, “I’m glad you think Google is great, but I think it sucks.”

His point, he quickly added, is that the objective of a search engine is to enable the reader to find what he or she wants. Google, Page noted, rarely does this perfectly. Even the simplest searches retrieve pages of choices, many of them quite beside the point.

There is no question that search generally, and Google (the dominant search provider) specifically, are modern miracles. But it’s worth noting that Google provides a button labeled “I’m feeling lucky” to take users to just one page in response to their search — and that a few years ago it revealed that only about one search in one hundred takes advantage of this option. Why? Presumably because users have found that the page often won’t be the one they were looking for.

Supposedly in response to this problem, search engine optimization (SEO) has grown up in the last decade as the dark art of online publishing. Sites hire SEO experts, divine new SEO practices, invest ever greater resources in SEO. SEO is composed of a number of techniques, some simple, others devilishly complex, for attracting the attention of search engines to content.

Some of these are innocent, like “tagging” content to indicate its subject. Some alter the content a bit, making it less variant and interesting, but probably not less valuable. Witty headlines are abandoned in favor of prosaic titles, returning headline practices to those of 19th-century newspapers. Sentences are shortened and standardized. But other effects are more troubling. “Content farms” grow up and harvest billion-dollar stock market capitalizations from creating content that search engines will serve up ahead of others — not, by definition, content that doesn’t already exist, but rather content that will enjoy a commercial advantage over other similar content that does exist but is less easily found. Websites profit and grow from appropriating content from other sites, and aggregating it in a way that will induce search engines to send traffic to the often superficial aggregator ahead of the more substantive original content creator.

But all of these practices and trends prove, more than anything, not that search rules the web, but rather that Larry Page was right, and that search engines remain deeply imperfect. That is because SEO, when reduced to its essence, is aimed at directing searchers not to the information they seek, but to where the publisher wants them to go; any overlap of the two is coincidental. Remember, it’s called “search engine optimization,” not “search optimization.”

And now a backlash is building that seems destined to have powerful effects. Three years ago, Jeff Jarvis titled a post “The end of SEO?” and asked if search was about to be overtaken by social media. Last year, Washington Post columnist Gene Weingarten wrote a column mocking SEO practices, the online version of which was headed “Gene Weingarten column mentions Lady Gaga.” The Columbia Journalism Review recently followed with “CJR Column Mentions The Simpsons.” Without any sense of irony, the Huffington Post, on the day it announced its merger with AOL and with the Egyptian revolution raging, devoted a lead story (and headline) to a classic SEO gambit: “What time does the Super Bowl start?”

Amid all this fun, however, the folks at Google seem a bit concerned, and are actually trying to do something about it. They’ve announced a major revision of their search algorithm to downgrade low-quality sites and the worst SEO game-players (including some content farms and retail giants like JCPenney).

But this is likely just a small step in an inevitable direction, with the direction being the sunset of SEO.

Meanwhile, SEO experts debate endlessly among themselves the differences between “white hat” SEO (manipulating content to make it more machine-readable) and “black hat” SEO (which tries to mislead the search engine with stunts such as thousands of links from sites that themselves offer nothing of value). What this entire debate misses, of course, is that SEO itself is an inefficiency, a transaction cost rather than a value-creator — it is a technique designed entirely to compensate for the failure of the search engine to correctly analyze site content, searcher desire, or both. Over time, economics teaches us, inefficiencies tend to be wrung out, and transaction costs reduced.

The types of content available to online users are growing — things can be done with data and with video that were not previously possible. But most of what people want to know at the moment they search is already known and has already been published. Unlike other “games” between an “offense” and “defense” — such as military technology or football formations, where each new development in “offense” can beget one in “defense” — the “game” between searchers and content will likely one day largely end. Technology will be developed so that searchers can find the content they actually want, quickly, easily, correctly — and at the highest level of quality available at whatever cost the user is willing to pay, maybe nothing, maybe something — without thousands of false answers or copycat answers or answers in drag.

Perhaps such a solution will prove too costly to develop, or to deploy. But Google has hundreds of our brightest young minds working on this problem, and they are well aware that if they don’t continue to make progress on it, someone else will — maybe the hundreds working on the same problems at Microsoft, maybe two kids as old today as Larry Page and Sergey Brin used to be.

The Huffington Post/AOL deal may mark something of a watershed in this progression. Much of the $300-million-plus in value HuffPo has built has been in playing very smartly by the SEO rules of the first decade of this century. But if it is true that most entrepreneurs sell out near the top, and it is, then perhaps we have just been sent a signal by one of its masters that the dark arts of SEO have peaked and that the century’s second decade will see them fade, perhaps into near nothingness by the third decade. In other words, it seems increasingly likely that, when the history of this era is written, SEO will turn out to have been a transitional phenomenon.

Some, starting from at least that Jeff Jarvis post of three long years ago, argue that social media will overtake SEO. Well, yes and no.

“Yes” in the sense that social media can provide a perfect search technique for some purposes. If you want a trusted recommendation on almost any commonly used product or service, your Facebook friends, if you have enough of them, can tautologically provide one. Tautologically, because you have presumably chosen them precisely because you trust them. And social media can provide a serendipity that search lacks, as your friends can identify answers to questions you hadn’t even formed yet.

But also “no” because, if the term “friend” is to remain meaningful (and, just as important, their number manageable), you don’t have enough friends for them to know the answers to all of the questions you may want to pose.

In any event, why does all of this matter? It matters most of all because the ability of content to attract readers is at the heart of nearly every web publishing model.

Web advertising was once heralded as a modern miracle, the answer to retailer John Wanamaker’s complaint that half of his (print) advertising dollars were wasted, but that he didn’t know which half. On the web, it was said, the wasted half was revealed to be more like 90 percent — but the advertiserdid know which part was wasted. To a significant extent, this has proved true, and online advertising has been a boon — to advertisers. But with the barriers to the production of content reduced nearly to zero, so much content has been produced, so much supply of places to advertise has been created, that, with demand for such space limited, online advertising prices have crashed. The result: For about five years now it has been increasingly clear that advertising will not be the linchpin of business models for publishing content in this century that it was in the last one. In one telling illustration, blogger Nate Silver recently calculated, to convincing effect, that the average Huffington Post blog post attracted about three or four dollars in advertising revenue.

Instead of advertising, content businesses, if they are to be successful at all online, will need to recover the cost of content creation increasingly from readers. (This is why the entire publishing world is about to hold its collective breath as The New York Times rolls out its metered pay plan.) And for the cost to be recovered from readers, the readers will have to be able to find the content. (Most publishers don’t begin with a brand like The New York Times.)

What does this have to do with the decline of SEO? A great deal, I think. SEO has been, more than anything, about growing pageviews and unique visitors — any pageviews, and any unique visitors, the more the merrier. It is a force, therefore, for lowest-common-denominator publishing. And after a decade of SEO, a lot of lowest common denominator is what we have.

But a focus on readers rather than advertisers as the heart of business model will, inevitably, create a more segmented dynamic, as the strongest appeals to readers tend to be in niches, and as, to venture an impolite reminder, some readers are a great deal more valuable than others. This is not only because some readers have more money to spend on content (as they do, admittedly, on the goods and services offered by advertisers), although that is true. But it is also, and ultimately more importantly true, that some readers are willing to spend more time, to develop greater loyalty to particular content, to value it more highly.

Improved search, and diminished SEO, should tend place a greater value on such readers, elevating content of higher value, higher quality and, therefore, higher cost.

That would matter a great deal.

POSTED     March 9, 2011, 1 p.m.
Show tags
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
Postcards and laundromat visits: The Texas Tribune audience team experiments with IRL distribution
As social platforms falter for news, a number of nonprofit outlets are rethinking distribution for impact and in-person engagement.
Radio Ambulante launches its own record label as a home for its podcast’s original music
“So much of podcast music is background, feels like filler sometimes, but with our composers, it never is.”
How uncritical news coverage feeds the AI hype machine
“The coverage tends to be led by industry sources and often takes claims about what the technology can and can’t do, and might be able to do in the future, at face value in ways that contribute to the hype cycle.”