“We celebrate our failures,” Google CEO Eric Schmidt said yesterday when speaking at the Techonomy confernce, in response to the surprise closure of his company’s Google Wave product. When it comes to failures, Google’s celebrating more than you might realize.
Some believe that anything Google touches is golden. Yesterday’s closure of Google Wave is another reminder of how this isn’t so. Below, a summary of important Google products that haven’t made the cut, over time.
For each product, I’ve also pulled a “celebratory failure quote.” I don’t mean for that to be as snarky as it seems. It’s meant to illustrate the difference between how Schmidt’s statement sounds and what his company actually tells the world.
I agree. Google’s a company that’s not afraid to take risks and does seem to embrace the idea that along the way, there will be failures. Maybe that’s “celebrating” those failures. But in its statements to the world, Google rarely sounds like it’s celebrating these missteps. It doesn’t really document anything that was learned. It just seems to say as little as possible to move on.
On to the failures, listed in order of closure date:
Google Wave (May 2009 to August 2010)
Google Wave was perhaps one of the most heavily hyped products that Google’s put out, only to have it fall on its face. The product was launched in a great splash (sorry, Wave opens itself to puns, intentional or not) at the Google I/O conference in 2009 (see Live Blogging Google Wave).
But rather than the revolutionary communications change that Google suggested Wave would be — a step beyond email and instant messaging — Google Wave found little adoption and lots of head scratching about what to do with it. The company announced it would close yesterday (see Google Wave Crashes).
In celebration of Google Wave’s closure, the company said:
But despite these wins, and numerous loyal fans, Wave has not seen the user adoption we would have liked. We don’t plan to continue developing Wave as a standalone product, but we will maintain the site at least through the end of the year and extend the technology for use in other Google projects…..
Wave has taught us a lot, and we are proud of the team for the ways in which they have pushed the boundaries of computer science. We are excited about what they will develop next as we continue to create innovations with the potential to advance technology and the wider web.
Google SearchWiki (November 2008 to March 2010)
SearchWiki allowed anyone to shape their search results manually. You could delete results you didn’t like and move others to the top of the list.
The celebratory quote:
Stars in search replace SearchWiki. In our testing, we learned that people really liked the idea of marking a website for future reference, but they didn’t like changing the order of Google’s organic search results
Google Audio Ads (January 2006 to February 2009)
In January 2006, Google acquired dMarc Broadcasting, a radio-based advertising platform, for $102 million in cash along with future performance payments that could have totaled up to $1.1 billion.
Google promised that it would bring to radio the type of measurable, performance-based ads that it was known for in search. A new Google Audio Ads product was introduced (logo above from the Natural Search Blog). Three years later, Google jumped out of broadcast radio ads. It turned out that Google couldn’t measure performance as it hoped.
There were other issues, including how Google worked — or didn’t work well — with radio stations. Radio Tunes Out Google in Rare Miss for Web Titan is a nice story about the failure, from the Wall Street Journal. Our summary of it is here. The New York Times also covered the early departure of dMarc’s founders in 2007: Google Encounters Hurdles in Selling Radio Advertising.
The failure celebration quote:
At Google we’ve never shied away from high-risk, high-reward projects. We believe that making big bets is not only in the best interests of our users and partners, but also important for our long term success. In 2006, we launched Google Audio Ads and Google Radio Automation to create a new revenue stream for broadcast radio, produce more relevant advertising for listeners and streamline the buying and selling of radio ads. While we’ve devoted substantial resources to developing these products and learned a lot along the way, we haven’t had the impact we hoped for.
So we have decided to exit the broadcast radio business and focus our efforts in online streaming audio….
We have always accepted that if you take risks not all of them will pay off. Deciding to close products is never easy, but we will continue to focus on advertising products that provide measurability for advertisers, and are relevant and useful for users, listeners and viewers.
Google Video (January 2005 to January 2009)
Wait, isn’t Google Video still going? You’re thinking of Google Videos — plural — which is what Google Video morphed into. You see, back before Google bought YouTube in 2006, it was already trying to build its own video sharing service. That was Google Video, launched in January 2005.
At first, Google Video ironically had no video at all, much less sharing. Instead, the service recorded TV broadcasts and turned them into searchable transcripts. It was actually pretty cool, though it apparently gave broadcast networks fits. But by the middle of the 2005, Google began allowing video uploads and sharing. You know, like YouTube. By January 2006, it also quietly shuttered the TV search part of the site.
Despite allowing uploads, Google effectively threw in the video sharing towel by purchasing YouTube, in an estimated $1.65 billion stock deal. The purchase was an expensive admission that Google couldn’t build on its own what YouTube had achieved.
But the failure of Google Video wasn’t yet done. In 2006, Google Video also rolled out an online video purchase and rental service. It closed that abruptly in August 2007 and offered $5 in Google Checkout credit to customers. That didn’t please some, so the company quickly offered full refunds.
Also in 2007, Google Video started morphing into Google Videos, a new “meta” search engine that found video content from Google Video, YouTube and video from across the web. Google signaled this would happen early in the year, then as part of its “Universal Search” rollout later that year, the meta search went live, even if it remained being called “Google Video.”
I’m not exactly sure when the “Google Videos” moniker started. Google Video still does exist to host video that was uploaded through January 2009, when uploads were turned off (see Google Ends Google Video Uploads, Shutters Notebook, Catalog Search, Dodgeball & Jaiku) — the effective last nail in Google Video’s coffin.
Any celebratory quote? Google’s press release about buying YouTube didn’t celebrate having to spent over $1 billion because Google Video couldn’t do what the company hoped. It said nothing about Google Video at all. As for the closing of uploads to Google Video, Google said:
At Google, we like to launch early, launch often, and to iterate our products. Occasionally, this means we have to re-evaluate our efforts and make difficult decisions to be sure we focus on products that make the most sense for our users.
In a few months, we will discontinue support for uploads to Google Video….
There are still great options for people who want to upload content to Google, and we invite them to explore YouTube’s dynamic global community or Picasa Web Albums. If you have questions or need more information please read our FAQ page.
We’re confident this decision is the right one for our users, and we’re looking forward to making Google Video an even better place for you to search and find videos from all over the web.
Dodgeball (May 2005 to January 2009)
January 2009 was a major month of celebration at Google. At the same time it celebrated shuttering Google Video, because it ultimately had to buy success via YouTube, it also announced another failure: Dodgeball.
Dodgeball was purchased in May 2005 (see here on Engadget, where the logo above also comes from). Some feel it could have turned into a Foursquare-like service for Google, and that’s where the irony is rich.
Dodgeball was cofounded by Dennis Crowley, who left Google in April 2007, saying “the whole experience was incredibly frustrating for us.” Crowley went on to start Foursquare, currently the hot location “check-in” service that just received another round of investment.
In fairness, I’ve heard Crowley saw on several occasions that Dodgeball was hard for people to understand and perhaps too early for the mobile devices out there. Then again, if Google had kept going with Dodgeball, perhaps it wouldn’t now have to try and push Google Latitude as its catch-up product.
Google’s quote to celebrate the Dodgeball failure:
Some of you may also be familiar with Dodgeball.com, a mobile social networking service that lets you share your location with friends via text message. We have decided to discontinue Dodgeball.com in the next couple of months, after which this service will no longer be available. We will communicate the exact time-frame shortly.
Postscript: Since I wrote this, TechCrunch has just posted a nice piece where Schmidt talked specifically about Dodgeball and what could have been: Perhaps Not Fondly, Google’s Schmidt Remembers Dodgeball “Quite Well”.
Jaiku (October 2007 to January 2009)
Some think Jaiku could have been Google’s Twitter. Google purchased the microblogging service in October 2007 and did little with it. By January 2009, it was closed along with several other products. The service still exists, but Google doesn’t actively support it.
Somewhat related is how Google lost Ev Williams back in 2004. Williams was one of the founders of Blogger, which Google acquired in February 2003. A year later, Williams left. Unlike with the Dodgeball founders, Williams was extremely conciliatory and in fact said:
I’m honored to have been a part of Google for such a historic period. If I was going to work for anyone, I’d work for Google.
And yet, I’ve always heard rumblings that the Blogger team never felt that well treated at Google. Rumblings aside, Williams got away. Google lost talent that went on to make another major hit: Twitter, which Williams cofounded and currently serves at CEO.
Speaking of Twitter and lost talent, Google also lost the current Twitter COO, Dick Costolo. Google gained Costolo when it acquired the start-up he oversaw, FeedBurner, in June 2007. Costolo left Google and joined Twitter in September 2009.
Again with the rumblings, my understanding is that Costolo wasn’t particularly happy with Google or how it dealt with FeedBurner. And again, rumblings aside, it doesn’t matter. Costolo’s widely acknowledged as a major talent, one that Google couldn’t hold on to.
The Jaiki celebratory failure quote:
Google has long believed that thoughtful iteration is the best way to build useful products for our users. As part of that process, we are always looking for ways to better focus our teams on the products that can have the most impact….
While Google will no longer actively develop the Jaiku codebase, the service itself will live on thanks to a dedicated and passionate volunteer team of Googlers.
Google Notebook (May 2006 – January 2009)
Another January 2009 closure was Google Notebook, a way to clip text, images or search results and save them in online “notebooks” that could be shared with others. Google decided to close that in January 2009. Ironically, one of the alternatives it suggested as “being actively improved” was the aforementioned SearchWiki, which closed the following year.
The closure quote:
At Google, we’re constantly working to innovate and improve our products so people can easily find and manage information. At times though, we have to decide where to focus our efforts and which technologies we expect will yield the most benefit to users in the long run.
Starting next week, we plan to stop active development on Google Notebook. This means we’ll no longer be adding features or offer Notebook for new users.
Google Catalogs (December 2001 to January 2009)
Touted as a way to search through consumer catalogs, Google Catalogs (logo above from Seeklogo) was launched with promise at the end of 2001 and soon virtually forgotten by everyone, including Google, it seemed. It was updated maybe once or twice, as far as I can recall, soon after it launched. Then it just sat. Google finally put it out of its misery in January 2009.
Google celebrated the failure by saying:
It was a great experiment. Nonetheless, in recent years, Catalog Search hasn’t been as popular as some of our other products. So tomorrow, we’re bidding it a fond farewell and focusing our efforts to bring more and more types of offline information such as magazines, newspapers and of course, books, online
Google Print Ads (November 2006 to January 2009)
Just as Google couldn’t transform radio ads, the company also failed to change newspaper advertising. Google Print Ads, launched in November 2006 (and tested as far back as mid-2005) put ads into newspapers. But the product apparently failed to generate much revenue or interest, so it was closed in January 2009. You can still see how they worked in this demo Google has left online, for now.
The celebration of failure quotation:
In the last few months, we’ve been taking a long, hard look at all the things we are doing to ensure we are investing our resources in the projects that will have the biggest impact for our users and partners. While we hoped that Print Ads would create a new revenue stream for newspapers and produce more relevant advertising for consumers, the product has not created the impact that we — or our partners — wanted. As a result, we will stop offering Print Ads on February 28….
We will continue to devote a team of people to look at how we can help newspaper companies. It is clear that the current Print Ads product is not the right solution, so we are freeing up those resources to try to come up with new and innovative online solutions that will have a meaningful impact for users, advertisers and publishers.
It’s always difficult to say goodbye to products. Lots of people at Google have worked hard on Google Print Ads. Some advertisers have seen good results and our partners have dedicated time and resources to help get it off the ground. But as we grow, it is important that we focus on products that can benefit the most people and solve the most important problems. By moving resources away from projects that aren’t having the impact we want, we can refocus our efforts on those that will delight millions of users.
Google Page Creator (April 2006 to August 2008)
Did the web need yet another way to build web pages? Google thought so and rolled out Google Pages in April 2006, with an official blog post. Two years later, Google decided that its Google Sites product was the way to go. Google Page Creator was closed without fanfare. There was no official blog post about it. For the celebratory failure quote, I have to cite what Google Blogoscoped pulled off the Google Page Creator “About” page and a help thread, at the time:
We are no longer accepting new sign-ups for Page Creator because we have shifted our focus to developing Google Sites, which offers many of the capabilities of Page Creator along with new features like site-level navigation, site-level headers, control over who can see and edit your site, and rich embeddings like calendars, videos, and Google docs.
Page Creator has always been a Google Labs project (Labs is our “technology playground” where we let users test-drive experimental products and give us feedback so we can innovate more quickly). Since launching Page Creator in Labs, we’ve learned a lot and have incorporated those lessons into Google Sites. We think the Labs program, which allows users to try new things before they’re fully baked, lets us innovate faster and ultimately create the best possible products.
Google Answers (April 2002 to November 2006)
Facebook’s just rolled out Facebook Questions, the latest in a long line of “answer” based search products from companies including Yahoo. But Google Answers, a Q&A service launched in April 2002, was killed off in November 2006. The irony is that year marked a resurgence in Q&A-based services, with the new Yahoo Answers gaining much attention and traffic. Unlike most services, Google Answers relied on paid researchers. But rather than reshape the product, Google axed it.
The celebratory failure statement:
Google is a company fueled by innovation, which to us means trying lots of new things all the time — and sometimes it means reconsidering our goals for a product. Later this week, we will stop accepting new questions in Google Answers….
Google Answers was a great experiment which provided us with a lot of material for developing future products to serve our users. We’ll continue to look for new ways to improve the search experience and to connect people to the information they want.
Could These Be Next?
Which products might face a failure celebration next. Some thoughts:
Orkut: Back before Facebook, Google was already playing the social networking game with Orkut, launched in January 2004. Over six years later, Orkut is big in Brazil, India and Iran (if I recall correctly). That’s pretty much it. Even in Brazil and India, Facebook may be edging it out.
Google Knol: Google’s semi-challenger to Wikipedia doesn’t appear to have gained much traction since it was formally launched in July 2008.
Google Sidewiki: Google’s tool that allows anyone to add comments to any web site appears to be little used. It was launched in September 2009.
For all that pain, Buzz seems to have had little adoption. Google’s touted no particular growth figures. Publishers across the web have uniformly failed to report any great traffic gains.
There are people on Buzz, but there were people on Wave. The question is, are there enough people for it to keep going. Potentially, Buzz will get rolled into the rumored “Google Me” social product.
From Failure To Success
This post has dwelt on Google’s failures because that’s the talking point of today, in the wake of Google Wave’s closure (honestly, no pun was intended — it just lends itself to puns in all sorts of shiny ways).
Google has had plenty of successes, huge ones, from its original search engine through to products like Google Maps, Gmail and the Android mobile operating system. Who’d have thought two years ago that Google would back an operating system that could threaten Apple’s dominance, much less RIM’s and Microsoft’s, in the smartphone space. Yet today, it does.
Certainly Google’s encouraged a culture of “20% time,” where engineers (at least — ordinary Googlers don’t seem to get this) can spend 20% of their work time on whatever they’d like. It does go off in various directions and clearly isn’t afraid of killing projects that don’t pan out well. That is embracing the possibility of failure, sure.
And sometimes those failures even come back as success. Case in point. Google Voice Search was introduced in 2002 as a way to call Google on phone, say your search and get back results. You can see an archived example of this here.
Google never formally announced the closure of Google Voice Search. At some point, the system’s home page stayed up, but nothing worked. It seemed forgotten. But years later, Google started moving ahead with voice recognition technology. Today, it’s built into a variety of Google services — and it’s a key component of Google’s Android operating system.
Postscript: A couple of people have pointed out some other notable failures. Lively is a good one, Google’s “virtual world” service that was closed in Nov. 2008. The Google Nexus One is another one. Launched in January 2010, with the promise of radically changing how mobile phones were sold, Google canned the effort after five months, closing it in May 2010.