The Big Money - Most ReadArticles

Google Does Non-Evil Thing: Bans White Teeth, Flat Stomachs

chadwick.matlin — Tue, 17 Nov 2009 18:06:56 +0000

The search giant gets it right in its newest anti-spam drive.

If you’re going to be the market leader, you might as well lead by example. Google (GOOG) has long dominated the online ad industry, fueled by its popularity in search and its auction-based approach. Ads on the Web, meanwhile, have been overrun by shady offers of teeth whiteners and stomach flatteners. These ads—which TBM has covered zealously—are classic bait-and-switches. They lure you in with the promise of self-improvement, charge your credit card exorbitant and unexpected fees, and don’t pick up the phone when you call to cancel. As I said, these guys are the salt of the earth. Unsurprisingly, Google’s ad network has not been immune to these scams despite their efforts. Ubiquity, after all, attracts the unseemly.

Google has long tried to weed out these ads and their scammer brethren. Its ad network is policed by software that detects the bad seeds and puts them on notice. (This is very different from some ad networks, like Pulse360 and AdBlade, that allow these scammers to spread their message across the Internet, including on The Big Money.) But the scammers are just as crafty, consistently tweaking their pages to sneak past Google’s requirements. It’s the oddest of arms races: Google trying to prevent some customers from giving it money, and the customers trying as hard as possible to make sure they can keep paying Google. When Google banned an ad and the site it linked to, the scammers just created a new domain and ad, essentially pointing users to the same place, filled with the same shoddy offers. Google’s old approach, while partially effective, still let too many ads through.

And so Google has made a minor shift in its policy that has major implications. Up until now it has taken action against ads, not advertisers. If an ad violated one of Google’s terms of use, the search giant would take it out of circulation, but that’s it. Google briefed TBM on its new policy: It will now ban the advertiser, not the ad, effectively neutering the advertiser’s ability to shift from one ad and shell site to another. Think of it like the struggle between the police and a graffiti vandal. Up until now Google has only been erasing the tags after they’ve been put up. Going forward, they’re going to take away his spray cans and put a GPS collar on him, making sure he never does it again. It would be a principled stand by any company, but especially by Google because of its position in the market. I worry, though, that the rest of the industry won’t pay attention. On this issue, Google might be a leader without any followers.

Google’s policy switch is so impressive because it reacts to the real-world circumstances of how these scams are perpetrated. The crackdown is meant to focus on all scams—malware, get-rich-quick sites, getting users to pay for otherwise free software, etc. But as a case study, let’s look at how it affects the white-teeth and stomach-flattener scams that pervade Internet ad slots.

To do so, I need to give a quick crash course in affiliate economics. If you’re a real masochist and want more information, I’ve outlined the whole sordid ecosystem elsewhere on TBM. The abridged version: These ads propagate through a network of affiliates that are paid to direct people to the scam product in question. Let’s say you’re a scumbag who wants to sell an acai-berry weight-loss regime. You’re not the one doing the advertising on search engines and content sites; you’re paying others to do it for you. The affiliates are usually directing people to an intermediary page that speaks highly of the scam product, making somebody more likely to buy if he clicks through to the product page. The affiliate then gets a cut of the revenue if the people it sends to your scam site stay and buy something. It’s the affiliates who are flooding the Internet with ads—the more they put up, the more likely they are to get clicks, referrals, and revenue. As long as they’re making more in commission than they are spending on advertising, they come out ahead.

Thus, it’s part of the affiliate’s business model to create as many ads as possible. And here’s where we get to Google’s common-sense adjustment. By targeting the affiliates, not the ads, Google avoids playing Whack-a-Mole and gets to the root of the problem. In the past, when Google banned an ad or linking site, the advertiser would just put up a different ad that linked to a cloned page. Google’s new policy will stop that. The company also says its software can detect if a scammer sets up a duplicate account in response to the banning, eliminating that new account, too.

(There are many concerned that Google will be too strict in its permanent bans, eliminating legitimate advertisers without explanation. Google obviously says that won’t be the case and that there’s a review process for those who feel they’ve been wrongly banned. As with the fight against spam, the key will be to limit false-positives.)

Google, of course, is not the only ad network that airs these ads. The major offenders are networks like Pulse360 and AdBlade, who are unabashed in giving affiliates a mouthpiece for their link-heavy intermediary pages. And because allowing these scams to continue isn’t illegal, there’s little incentive for ad networks to follow Google’s lead. There’s just too much money at stake, especially when scammers flee to other networks after Google ferrets them out. AdBlade is now serving ads to 150 million unique viewers a month, and much of their inventory is taken up by, how shall we say, non-marquee advertisers. Google has the luxury to crack down; many smaller ad networks do not. Economic incentives don’t encourage good behavior.

And so we’ll have to wait and see whether others follow Google’s lead. Until then, at least somebody has acknowledged the mechanics of the marketplace. There’s no doubt the ad-network community will hear the message. The question is whether anyone will listen.

TAP Tagline:

Google Does Non-Evil Thing: Bans White-Teeth, Flat-Stomach Scams

The Trouble With TripAdvisor

caitlin.mcdevitt — Thu, 19 Nov 2009 19:40:29 +0000

There are good reasons not to trust crowdsourced sites.

If you’ve searched online for a hotel recently, chances are you’ve stumbled upon TripAdvisor, the hotel review Web site. TripAdvisor ranks hotels according to crowdsourced feedback. Anyone can go to the site to rave or rant about a stay in a particular hotel. If the toilets at a certain hotel don’t flush, you can let the world know on TripAdvisor. If you’re charged extra for bacon with your breakfast or the toast is not warm enough to melt butter, you can warn other travelers to steer clear. Or if you had a wonderful stay somewhere, you can write about it in glowing detail. In fact, most of the reviews on TripAdvisor are positive.

When it popped up in 2000, TripAdvisor recognized a huge information gap between travelers and hotels. Hotel Web sites only showcased flattering pictures and sometimes too-good-to-be-true descriptions. Travel agents were expensive. And traditional hotel rating systems—typified by stars and diamonds—were widely recognizable, but few people really understood what they meant. (Most people still don’t.) TripAdvisor positioned itself as the go-to spot for straightforward and detached criticism courtesy of the traveling masses. Like Yelp and Wikipedia, TripAdvisor eschews institutional expertise in favor of the wisdom of crowds. All three Web sites flourished and grew because the public deemed them useful, and as more people used them, their authority grew.

TripAdvisor, now owned by Expedia (EXPE), took off, but—perhaps thanks to its high profile—it’s becoming yet another case study in crowdsourcing headaches. Just as people have been nabbed editing their own Wikipedia profiles, inflating their Amazon (AMZN) book rankings, and Yelp-ing false reviews about rival pizza shops, there are people who manipulate their popularity on TripAdvisor. Obviously, it’s the hotels that have the most to gain or lose from high or low ratings. They’d be foolish to ignore the site—which gets 36 million monthly visitors—and, some may say, foolish not to do what they can to improve their rankings. This past summer, TripAdvisor admitted that it was having trouble with hotel staffers posting reviews of their own properties. The company assured everyone that it has a dedicated staff who can sniff out fishy posts. It also slaps a red warning label on the profiles of any properties that it suspects have manipulated their own reviews.

Still, there’s not much that TripAdvisor can do to guarantee its credibility. Because reviewers can be anonymous, it’s not difficult for a crafty hotel marketing officer to commission underlings to post positive reviews of a property on the site. Or they can ask family and friends to do so. TripAdvisor’s top-ranked property in Miami Beach right now is a boutique hotel called the Betsy. It beat out 195 other reviewed properties for the top spot. But as one TripAdvisor reviewer points out, “I would note that we were told when we checked in that we were some of the first paying guests as march and April they had been trialling with friends and guests of the owners etc. This possibly explains why none of the reviews posted so far mentions the fact that the hotel is not completed.” The hotel management responded to this post defensively, claiming that all reviews came after this trial period, implying that friends and family had not written them. That may well be true, but, frustratingly, there’s no way for TripAdvisor readers to know who to trust. There’s also no built-in mechanism to make sure reviewers of a certain hotel have actually stayed there, which welcomes even sneakier ploys—including, apparently, reviews for hire. This now-expired job posting on Scriptlance.com, a freelance job board, offers cash for 100 TripAdvisor reviews of a certain hotel.

Crowdsourcing enthusiasts might claim that TripAdvisor’s value isn’t diminished by nefarious hotel marketers because their voices are ultimately drowned out by the rest of the community. But it seems like the rest of the industry isn’t so sure. Recently, Forbes took over licensing of Mobil Travel Guide’s old-school star rating guide and says it has plans to expand it. Oyster, a travel review Web site launched this summer, employs journalists trained as hotel critics to write reviews. Both ventures are betting that the final say on which hotels are best should fall to answerable professionals—who, before TripAdvisor, had always played that role. The pendulum seems to be swinging back that way, perhaps for good. Even if TripAdvisor is able to weed out self-promoters, a fundamental question remains: Are the rest of the 30 million reviews worth reading? That, of course, depends on who you ask.

Montezuma’s Aztek Revenge

matthew.debord — Tue, 17 Nov 2009 12:05:22 +0000

How the reviled Pontiac could lead GM into the future.

’Tis the season for new car models. Fall is when the automakers start rolling out their offerings for the next year. All-new models arrive, redesigns hit the dealerships, and updates to existing cars appear. But obviously, there’s a wrinkle this time round: Both General Motors and Chrysler are fresh off bankruptcy. And you’d be right to assume there’s extra pressure on their new wheels.

Chrysler, unfortunately, doesn’t have much new stuff to sell. But GM has a new SUV, the GMC Terrain, that is already a sales leader and recalls a vehicle that still horrifies carmakers and critics alike. That car is the despised Pontiac Aztek, an almost universally loathed vehicle that established the paradigm for the Terrain: the crossover SUV, a half-car, half-truck concoction that is one of the fastest-growing vehicle categories. GM needs to remember the Aztek, because it represents the kind of risk-taking design that the post-bankruptcy firm will need to go forward. The temptation for the New General will be to copy successful market formulas, rather than try to define new market segments.

The Aztek, introduced in 2001, was an attempt to do something entirely different. It was aimed at then-twenty- and thirtysomethings who liked to hike, camp, mountain bike, and generally participate in the whole suite of Outside magazine diversions, but who might also want a young-family hauler with a bit more flash than your typical truck or SUV. So the Aztek came furnished with a host of outdoorsy options, an interior that could be configured according to the recreational preferences of customers, and an all-wheel-drive system for the snow and the mud and the slush and the rain. The design was boldly idiosyncratic, but GM figured it would attract buyers. It wound up scaring them, but at least it took no prisoners.

It’s easy to berate GM for always failing to see where the market is going. But in this instance it was the first to recognize the need for a new kind of vehicle to fill the crossover segment, which would grow rapidly in subsequent years. A crossover is basically a 21^st-century station wagon. SUVs are usually built on the same platform used for trucks—and they often feel that way when you drive them. They also inhale gas. Crossovers, by contrast, are built on platforms used for cars, so they have better road manners, and they’re more fuel-efficient. There were some crossover-ish vehicles before the Aztek, such as the Subaru Forester, but these were seen as neo-wagons, or small/compact SUVs. With the Aztek, GM created something that had SUV size, minus the SUV stigma.

An innovative GM? Well, yes. GM can sometimes be, for all its detractors, troublingly ahead of the curve. And the Aztek was first in this mold. It was good at what it set out to do, despite the zany styling. And it showed that the four-door sedan, the hatchback, and the midsize SUV could be meshed. The Pontiac packaging was profoundly flawed, but the concept and engineering execution were solid. GM later rebadged it as the Buick Rendezvous and salvaged some sales before the product cycle petered out (the Rendezvous was much better received by families who wanted a more polished, less aggressively styled car).

In terms of innovation, the Aztek shares DNA with some surprising relatives, like Apple’s early, failed PDA, Newton, or its first stab at a portable, proto-laptop Mac. Apple (AAPL) didn’t succeed with these products, but the company began to define new markets with them. Obviously, laptops and notebooks would eventually become huge part of Apple’s business, and while Palm came to dominate the PDA market, Apple’s experience with Newton set the stage for its move into smaller personal devices, such as the iPod and iPhone. GM could banish all recollection of the Aztek, but the vehicle’s controversial design could be just the ticket as GM seeks to define how hybrid gas-electric-crossover technology derived from the Chevy Volt will appear.

Still, for anyone born in the 1980s or 1990s, the Aztek is increasingly the new Edsel, Ford's infamous automotive failure from the late 1950s. This would incline a swath of GM designers and engineers not to dare utter its sullied name. They don’t like the rather strange front fascia, nor do they care for the elevated rear end. They don’t like the lower-body cladding. But all these aesthetic objections are misplaced. The Aztek didn’t work, but it demonstrated that GM had the capacity to invent a product that people didn’t know they wanted. The General can still do this—the forthcoming Chevy Volt-extended-range electric car could be a game-changer for the company. But it needs to keep doing it.

And even though it might fail miserably … well, that’s the auto industry. Success is never guaranteed. But blandly hewing to what has worked, falling victim to fear rather than having the confidence to completely miss the mark from time to time, will not bring GM back to its glory days. Or even, someday, enable the company to return to profitability and pay back the taxpayer. So remember the Aztek. It may not have been great. But it gave birth to a new idea in the auto business, and that’s gold.

TAP4 Image:

Divorce, Reddit-Style

paul.smalera — Wed, 18 Nov 2009 21:09:16 +0000

How a social-media dream deal with Condé Nast fell apart.

In October 2006, Reddit co-founders Steve Huffman and Alexis Ohanian decided it was finally time to sell. A year earlier, their Digg-like social-news Web site had attracted an offer from Google, but the founders wanted to keep developing the fledgling company in-house. This time, though, the buyer that came calling was not the 800-pound gorilla of search engines. It was the 800-pound gorilla of magazine publishing, the one wearing Gucci loafers and a Patek Philippe watch: It was Condé Nast. Reddit was rumored to have sold for $65 million—a number that was denied by both sides. Reasonable analyses have put the price at $15 million to $20 million.

Fast-forward to today. Although Reddit has had solid traffic growth since the sale, both founders, now 26, announced they were leaving the company on the day their contracts with Condé Nast expired: Oct. 31. Halloween.

Granted, the marriage always seemed forced. But the dissolution of the Conde Nast-Reddit partnership says a lot about the oil-and-water mixture of old media and new. In an interview with Drew Schutte, the latest VP to manage Condé Nast’s online properties, just 10 days after Huffman and Ohanian’s exit, about all the company’s online initiatives, the word Reddit is never even used.

Three years ago, Condé Nast saw Reddit as a tool—one it could use to pry open the front door with other Web companies and one that would repair its own online track record, which was mixed at best. “We are confident that other companies will find Reddit to be a partner that can bring tremendous value to their Web efforts,” Steve Newhouse, the executive in charge of CondéNet, said back then. Huffman saw a chance for his company to grow, saying, “We’re excited … gain the resources to give us the freedom to accomplish many of the goals we’ve been dreaming about.”

Kourosh Karimkhany, the vice president of acquisitions for CondéNet who said he “played video games” with the founders while he was engineering the deal, said, “Our goal will be to build Reddit as an independent company by collaborating with Wired through the integration of its core technology, and by offering partnerships to allow others to do the same.”

That seemed sensible enough. And while Reddit was never a serious traffic challenger to Digg’s 40-million-plus unique monthly visitors—Ohanian says Reddit serves 7 million unique visitors a month; the highest independently available source puts the number at 5.5 million—the traffic nonetheless grew during the three years and ought to have been large enough to turn a profit.

Especially because, even under corporate ownership, Reddit kept a lean team. Digg now sports at least 75 employees, while Reddit has been getting by with “six and a half,” Ohanian told me, with some pride in his voice.

So what went wrong? For starters, the idea of Reddit partnering with other media properties never really panned out. One of the few that got built was the Independent, a British newspaper. The idea was to mix Reddit’s voting function—in which readers give a thumbs-up or thumbs-down on stories—with the Independent’s content. In an interview, the editorial director of the newspaper joked they wouldn’t ditch Reddit even if it didn’t deliver them “an audience of five million.” When I checked the site, there was only one link on the page with a double-digit vote count: 12 votes for a three-day-old story about drinking a bottle of wine daily. (For comparison, top stories on Reddit’s home page can hit 1,000+ votes in a matter of hours.)

Perhaps Condé Nast overestimated Reddit’s utility as a tool to build external Web site partnerships. After all, every media site worth its salt now has a row of those little social-media-sharing icons somewhere on its page, and they’re free and simple to install. But why not turn Reddit loose internally, not just on Wired.com, but across the Web sites of Condé Nast’s magazine empire, where The New Yorker, Vanity Fair, and Vogue stood just three years ago as temples of culture and power? By giving Reddit an exclusive foothold on the publishing company’s own content, Condé Nast would’ve at least been eating its own dog food. But Karimkhany seemed to really mean it when he said Reddit would be an independent company.

I wanted to ask Karimkhany about why Condé Nast shorted its multimillion-dollar purchase on staffing, even as Digg was growing by leaps and bounds. Karimkhany was also the general manager of Wired Digital, so he was starving the very puppy he brought home from the pet store. But Karimkhany, who was laid off in November 2008, and who appears to be working on his own startup, declined to talk and forwarded my e-mail to Condé Nast.

When I asked Ohanian why there had not been more integration in the three years he and Huffman were there, he said, “I really don’t know. We’re in a terrarium. Even stuff that happens outside my door is foreign to me.” It’s not hard to imagine the executives of 2006-era Condé Nast thinking they had very little to learn about media from a pair of 23-year-olds, even a pair they had just made millionaires.

When I asked Huffman, who, like Ohanian, is now 26, whether Condé Nast could’ve handled the acquisition differently, he demurred, writing, “Yes there were things Condé Nast could have done, but there were also plenty of things we could have done as well. Both sides have done some learning.”

Indeed, Reddit is a new chapter in an old story of mismanaged social media acquisitions. Take Del.icio.us, which was developed just a year earlier by a Wall Street quant named Joshua Schachter. Yahoo (YHOO) purchased the social-bookmarking site nine months before Condé Nast bought Reddit. Three-and-a-half years of turmoil later, Schachter also left his company, writing in to TechCrunch that, “I was largely sidelined by the decisions of my management. … It was an incredibly frustrating experience.” I had a meeting with Schachter during my time at (disclosure) Condé Nast Portfolio, where, even for an executive doing a dog and pony show, he seemed distracted.

Despite Yahoo’s tech roots, when Schachter joined it was already a company organized around the selling of advertising, much like Condé Nast and much unlike Google (GOOG), whose products all flow from job No. 1: providing the best search experience in the world. It’s not hard to see how a bookish guy like Schachter got lost in the shuffle at Yahoo, or how the same thing could happen to Huffman at Condé Nast. Both programmers, for example, needed a wakeup call about their Web site’s aesthetics. Proudly nerdy early versions of del.icio.us looked about as user-friendly as the interface of a Commodore 64. Even the URL, del.icio.us, was chosen by Schachter as an exercise in linguistic regression rather than usability—but an in-joke at a startup can look like a full-blown nerd alert at a mature company with a sales-driven culture. Which is why it’s shocking that Reddit has hung on to its similarly retro design, something that repeatedly stopped me from using it. It was only by overcoming the interface that I grasped Reddit’s great user base and functionality. (Huffman told me he found Web 2.0 trendiness to be “tacky,” which might explain why he didn’t embrace one good aspect of Web 2.0: usable, intuitive interfaces.)

I don’t know whether Ohanian or anyone at Condé Nast ever asked Huffman to change the site’s look, but someone should have, for starters. Then someone should’ve realized that they bought Reddit not just from Huffman and Ohanian, but from Y-Combinator, a venture capital firm specializing in seed-stage startups. Which means that Reddit was at best a midstage company in need of further incubation, guidance, and growth, not one ready to put in a “terrarium” and watch it grow. Reddit’s arrested development is symptomatic of other CondéNet acquisitions, like the once high-flying Ars Technica, which now seems starved under new ownership, deprived of sunlight, nutrients, and oxygen.

It’s a testament to the greatness of the concept that Reddit’s made it this far at all. Many other CondéNet projects not directly tied to magazine content, like Flip, stumbled almost right out of the gate. Even buying computers to run Reddit.com proved to be a struggle; after doing an end-around the IT department and expensing $40,000 worth of servers in the early days, Huffman found it was easier to switch Reddit over to Amazon’s (AMZN) Cloud Computing services rather than keep fighting battles over procurement. “We no longer have to bug Condé Nast for servers, and we no longer have to deal with ColoServe (the datacenter folks), who were absolutely horrible.”

With all the distractions, it’s no wonder the cofounders and faces of the company decided to check out once their contracts were up. (A third cofounder, programmer Chris Slowe, remained. The fourth, Aaron Swartz, was asked to leave* two months after Condé Nast bought Reddit.) In fact, after their departure announcement, Ohanian admitted they had been getting ready to leave for the last six months. Though Valleywag reported rumors of their departure two weeks before it happened, Ohanian was having a hard time hiding his plans. During a video interview with Andrew Warner of Mixergy back in June, leaving was already on his mind. “You’re a big success, you can get funding for any business you have,” says Warner. “Why are you still working for The Man?”

“It’s the coffee,” Ohanian tried to kid, before answering, “If it were just about the loot, it’d be a lot easier.” He then explains his attachment to the site and the “little freaking alien,” (Reddit’s mascot):

Given that everything here was pretty tolerable, we had no reason not to go, [laughs at his verbal slipup] we had no reason to go, we had no reason to go, because we were given plenty of freedom. This is my first job working for The Man where I didn’t have to ask for a bathroom break.

When Warner presses the question, saying freedom to use the bathroom and things being “tolerable” are pretty poor reasons for dot-com moguls to stick around a company, Ohanian says that “the very first moment I can say that I really, really dug it … was when I got to go up to the floor of The New Yorker, and they just let me wander around because I had a badge. And I was just wandering around the offices of The New Yorker and I was like, ‘This is pretty fucking great.’ ” Three years later, like countless other 23-year-olds who walked those halls thinking those thoughts, he’d be gone.

*An earlier version of this article incorrectly stated that Aaron Swartz never worked at Condé Nast.

TAP Tagline:

Why Condé Nast Execs Don't Even Say Reddit Anymore.

Presented By:

Wed, 18 Nov 2009 21:09:16 +0000