Put Google CEO Eric Schmidt on stage with SEW’s Danny Sullivan and you get an entertaining and enlightening glimpse into the soul of the world’s favorite search machine.
A special report from the Search Engine Strategies conference, August 7-10, 2006, San Jose, CA.
It’s become a cliche that keynote conversations at Search Engine Strategies are to “standing-room-only houses,” especially when a top Google exec is involved. Although attendance at Eric Schmidt’s keynote conversation with Danny Sullivan looked like another all time record, the seating problem was finally solved. There were enough seats for everyone in a cavernous hall at the San Jose convention center. Close to 3,000 eligible trade show attendees sat in on the session.
The first thing that was readily apparent, for anyone in doubt, is that Eric Schmidt really is a CEO. Many of his words were carefully chosen, yet sounded natural and unscripted. Many of his comments were asides that didn’t lean on prepared notes or rehearsal, yet none deviated far from the script. His technical acumen was a plus, but his ability to grasp strategy and grapple with an increasingly complex ecosystem allowed Sullivan to ask the tough questions without meeting a brick wall.
Not far into the conversation, I realized something else I hadn’t known until that moment: Eric Schmidt is my kind of guy. In a way that is bound to irritate spouses, laypeople, and “the rest of us,” and possibly get on the bad side of Alex Trebek, he’s my kind of guy.
That, based on an anecdote he told. A blustering politician told him authoritatively that “there are more outhouses than Tivos.” Said Schmidt: “Let’s look it up.” A perfect job for a Google Search, and the answer easily found within a few mouse clicks. As it turns out, growth in Tivo ownership had pushed total units past the number of outhouses. Looking it up, and being right, isn’t the most endearing trait, but it’s the only way to live if you want to build a global advertising powerhouse that runs on data rather than assertions. Or even if you want to know the right answers to questions. In Schmidt’s words, “it’s a great way to live.”
Sullivan set the theme for the conversation by asking Schmidt about his concept of “never betting against the Internet.” This led to some interesting background about Schmidt’s time in the area of client-server computing. That old business model supported a lot of proprietary protocols that required companies to hire lots of people to build and sell technology products.
The new model, according to Schmidt, is the idea of data on servers in a non-proprietary world that allows applications to be accessed and built on a universal platform, or what is sometimes referred to as “cloud computing.” Schmidt referred to the growth of advertising-supported models, but this seemed to overlook the development of newer-generation for-fee software that also doesn’t bet against the Internet (Freshbooks comes to mind). Indeed, I’ve been harping on the problem of assuming ads have to be the business model ever since Gator tried to bamboozle us into believing that adware was the only way to support a form-remembering utility. (At the time, I said I would have paid a fee. Now, it just seems redundant, as browsers will do it for free.)
The hot topic of the week—and since—has been AOL’s inadvertent leak of a large volume of search query data. Sullivan made it clear that the fallout was greater than anticipated: “anonymous” search streams were easily reconstructed by third parties. The New York Times found a woman who is now a “poster child” of Internet privacy fears: they called the 62-year-old widow from Georgia and asked her if she was one of those anonymous search query files, and she said yes. Referring to other sensitive search streams like apparent murder plots, illegal activity, and so on, Sullivan asked Schmidt what Google is doing to protect users against potential accidental or governmental intrusions into personally identifiable search details.
In the AOL case, it would have been prudent “not to release the information [to stray from normal security protocol to share it within the company” in the first place,” said Schmidt. In Google’s case, “We depend on our users,” he added. “The idea that an individual could be personally hurt by having data revealed about them is something that bothers us, so we have security systems to prevent it.” Moreover, “I’ve always worried that the query stream is fertile ground for government to snoop. That’s why it’s so important that a federal court judge made an impartial decision on Google’s favor recently [in response to Department of Justice requests for search query data”.”
Sullivan drilled down further to ask if search engines should filter information so personal details are harder to dig up using a simple search query. Schmidt’s response was detailed. He brought up the answer of a Department of Motor Vehicles policy that allowed personal information to be public. A side effect of that was that an actress was stalked and killed. Google is also working on things like prohibiting credit card numbers in its index, and making it easier to delete a phone number that appears online by directly contacting Google and asking for it to be removed from the index. Ultimately, it appears some editorial decisions need to get made in such hard cases, such as the placement of a site that lists abortion doctors and their addresses along with the statement “This is not a site to encourage you to kill them.”
Despite such hard decisions, Schmidt argued that “on balance, the trend towards making more information more widely available is a positive mission.”
Sullivan pushed Schmidt on the click fraud issue. Not satisfied with a recent Google announcement that it will disclose an estimated click fraud rate within accounts, Sullivan wanted to know when more detailed information will be available in the aggregate as well as for specific advertisers. While failing to go into specifics, Schmidt stated “I think we will eventually do everything you suggest.”
Sullivan also took issue with how much of a black box Google’s financials are, despite them being a large publicly-traded company. If Sullivan takes a special interest in the health of the search advertising market, he’d like to know more specifically about the revenue from just those ads—yet much of the reported information combines contextual ad revenue with search.
Surprisingly, Schmidt argued that such disclosure would help competitors “reverse-engineer every aspect of the model: CTR’s, CPC’s, RPM’s, and furthermore, based on incomplete information they may make simplifying assumptions are not true.” The unusual solution to that problem is to “not reveal the underlying economics of the ad box. That’s how we run the business, for numerous reasons.”
Turning to lighter topics, Sullivan asked what the new trends might be in placing ads in media like television, and whether that was the right focus for a search company.
Schmidt prefaced his answer by saying that this was the type of question the analysts tend to ask, and “we’re much more focused on the advertiser questions than the analyst questions.” Ultimately the principles are what are important. If advertisers need more ways of reaching audiences then they’ll try these new approaches. Google’s biggest challenge, Schmidt seemed to imply, would be simply to explain to advertisers the process for uploading and buying new kinds of media. Execs like Tim Armstrong are charged with selling the advertiser community on the benefits of new uses of the AdWords auction platform.
This report will continue in tomorrow’s SearchDay.
NOTE: Article links often change. In case of a bad link, use the publication’s search facility, which most have, and search for the headline.
From The SEW Blog…
- AOL Fires CTO & Two Employees After Search Records Slip Up
- ClickTracks Acquired By J.L. Halsey
- comScore Figures Show First Google Decline For Nearly A Year, But What To Believe?
- Subdomains Trademark Violations?
- Google Base Adds Click & Impression Metrics For Merchants
- Farecast Adds Cities To Flight Fare Prediction Tool
- My Q&A With Eric Schmidt In Video & Transcribed
- Aug. 21, 2006 Search News Forecast: Cloudy!
Headlines & News From Elsewhere
- 30 Days of Yahoo Publisher Network, Search Engine Journal
- Splogs Threaten Blogosphere, Wired Says, Micro Persuasion
- Authors@Google, Inside Google Book Search
- [Jason” Goldman is leaving the building [Google”, evhead
- Google Checkout’s Benjamin Ling Interviewed, AuctionBytes TV
- TrackMeNot Extension, Google Blogoscoped
- Cre8asite Forums Turns Four Years Old, Search Engine Roundtable
- Survey: Google Checkout Not Popular with Online Retailers, Google Watch
- Yahoo! homes in on Orange deal, The Guardian
- Yahoo quietly tests antiphishing service, IDG
- Matt Cutts Talks German SEO Spam, SEO Book
- Bypass SEM with Smart Website Strategy, iMedia Connection
- Sending Signals to Search Engines, ClickZ
- Google Groups Updates, Google Operating System
- Cairo office for Google, AME Info
- Is Google Today’s Microsoft?, Searchblog
- Search Rankings ‘Aint All That?, V7N
- Top Five Really Useful Free SEO Tools, V7N
- FeedCrier, another way to act fast with RSS and IM, TechCrunch
- Recent Trademark Filings: Yahoo, AOL, Viacom, paidContent.org
- Test Your Site’s Link Popularity with Socialmeter, Micro Persuasion
- Art Museums: A Collection of Searchable Databases #3, ResourceShelf
- The Search Battle For IE7, Google Operating System
- The Joy Of Vacation Rentals Searching On The Web (Not), Daggle