Shouldn't We Be Celebrating Google's Failures?


There has been an ocean’s worth of schadenfreude spilled over Google’s (s goog) decision to kill Wave, its experimental real-time collaboration tool. The reasons: It never had a purpose, it was too over-engineered in typical Googley fashion, it was a collection of features in search of a need, and so on. But shouldn’t we be celebrating the fact that Google was willing to experiment at all? That’s Google CEO Eric Schmidt’s view — he told reporters at the Techonomy conference that “we celebrate our failures,” saying the company encourages staff to take risks and possibly fail. And he is right to do so.

Yes, Wave was a hodgepodge of geeky features (hey look — I can see you typing while I’m typing!) without any real compelling use case. But the idea that people might want to collaborate in new ways isn’t a crazy one, and the concept of blending wiki-type features with instant messaging and document management isn’t really all that far-fetched either. It’s possible that Wave was just a little too early, and that most people haven’t even gotten used to the idea of Google Docs, let alone an all-in-one Swiss Army knife version. That doesn’t mean it was a stupid idea.

In any case, the impulse to experiment with such things and then release those features and services into the wild is arguably one of the best things about Google — such as the 20-percent projects that occasionally become Gmail or Google News or one of the company’s other big success stories. John Gruber of the Apple blog Daring Fireball writes that he is surprised Wave was even shipped, and that no other company would ever have released it. That may be true — and Apple certainly wouldn’t have, since its focus on hyper-designing everything borders on the pathological.

But that said, isn’t one of the positive things about web companies that they can experiment with things, and “fail quickly and fail often?” Isn’t that what so many startup advisors keep telling companies they should be prepared to do if they want to succeed? It’s kind of refreshing that a company the size of Google is still tossing experiments out into the world to see what happens to them and then try to learn from that. There aren’t many other large companies that are willing to do that.

So maybe Wave was poorly designed, or over-engineered, and didn’t deserve to live. But we shouldn’t be so quick to condemn Google for releasing it or experimenting with it — if anything, we should be cheering them on. Here’s a video of Eric Schmidt talking about Wave and other topics such as net neutrality at the Techonomy conference:

Post and thumbnail photos courtesy of Flickr user Iwan Wolkow


Jim H

Here’s what sucks about Google: the interfaces. Gmail is useable. Maps, great. As for the web apps, they’re either bland and boring or just hard to figure out, and illogical.

Example: I have Google Voice, on the Mac and the iPhone. Yes, it’s there, in stunning HTML 5! Now, I got the number, good; I connected my iPhone, my landline, and Skype. I got the SkypeIn number to use it. But Skype doesn’t work with Google Voice. I was stymied trying to fix my setup for about a week, until somebody read somewhere that it doesn’t work with Skype.

Their interfaces are a snap for techies. Not for the rest of us.


Wave, in some significant ways, seemed like an elegant grown-up version of Lotus Notes, but, like Lotus Notes, it tried to do too much and was brought down by its own weight. I certainly encourage Google to try and fail occasionally on the way to making better products, but sometimes public failing is incredibly damaging, and Apple’s model of failing in secret seems like the much better course.

Jack C

Exactly so!

In an evolving landscape, like the social web, industry must, among other things, dare to fail often if they hope to succeed in the long run. Pioneering is like throwing a Velcro ball at a target board. Ideas are only going to stick some of the time, so you can’t put everything into a single attempt.

Wave, by the way, didn’t fail because it was over-engineered, poorly designed, or too restrictive on invites. It failed because it not only didn’t play nice with other Google platforms, but because it was pitted in direct competition with different Google platforms that already had significant adoption (e.g., Docs, Gmail, Talk, etc.). Because of that, it never had a chance.

As always, Google desperately needs to make interoperability more of a priority. Wave should have been a projected that connected platforms, not one intended to supplant them.


Apple worked on iPad for 7 years before they released it. That is 7 years of failing to make an iPad before they succeeded in making an iPad. I celebrate the 7 years of failing because that’s how you get something great done, but I celebrate the success even more, because that is when they handed iPad over to me and now I’m doing great things with it in my own work.

With Google Wave, I don’t mind that it sucked when it was first launched, although I do think they launched it too early. What I mind is that they didn’t improve it, they didn’t fix it. They might have made it faster or more stable, but they didn’t give it a useful interface, they didn’t make it more practically useful for 99.9999999% of humanity.

So it isn’t just that Google releases a lot of half-assed products, it’s that they kill them before they fix them. How an I going to commit to a product like Wave if I know Google is very unlikely to?

Stark Ravin

Amen, amen, amen, amen! @Hamranhansenhansen has EXACTLY hit the point as to why Google sucks. They are just like Microsoft: a bunch of loser engineers who only know how to copy. The old adage about Microsoft getting a product right on version 3.0 is just like Google’s Beta tag. What it really means is that, like most engineers, they have no clue what they are doing, so they throw crap out there to see what beta users tell them about it. Then they follow them around like puppies hoping they’ll drop the next treat and let them know what to do next to refine the product. It is a timid, ball-less way to ship a product.

The utter truth is that Apple is the only tech company that knows how to conceive, build and ship a hardware or software product in this industry. And that’s because it is run by product people, not engineers. Engineers, by definition, only know how to do what they are told to do, either by the marketing team, or by copying others.

Another perfect Google example of this is Andy Rubin, another Google engineer-head who hasn’t had a creative idea in his life. If you look at early prototypes of the Android phones, they all looked like Blackberry devices. Then Apple, a TRUE innovator, ships the iPhone, and suddenly, all Android phones that ended up shipping become touchscreen devices that (poorly) ape the iPhone. Oy!

How Wave was handled was an effing abortion. Google blew it. They had something great and they didn’t develop it and stand by it. They are the 21st century Microsoft: a soulless company of software engineers who lean on others and their left brain for everything, rather than rely on the human spirit to make good product decisions. They are losers, one and all.


I’d say yes, people need to be free to experiment, to try things that might fail. But Google went wrong by not being able to see the failure it was before it actually hit the street. They should take a page from Apple’s book and keep things in-house until you’re sure it’s been tuned and trimmed of fat and needless features. It’s not done until you can’t think of anything else to remove without hurting the project.

Google seems to have taken the opposite approach and never put much critical thinking into “does this work?” before they let it out to the public. Buzz is another good example of not enough critical thinking before releasing it (and not even thinking about the privacy concerns of how it was executed).

So maybe Apple is too cautious – they too have a lot of failures, we just don’t hear them making a big fuss about it. But Google seems to be too uncritical on what they let out into the wild.


“without any real compelling use case” I don’t agree.
Wave was built as a way to manage the switch between sync and async communications that is needed by a team that is spread across multiple timezones where the working day only partially overlaps.
It did this rather well. (not least because it was built in Oz – serveral timezones away from the Googleplex)

It failed because :
1) Silicon valley is fantastically introspective and doesn’t care about anywhere else – so timezones are a null issue.
2) The pundits and hyper-geeks work weird hours anyway so don’t grok the problem.
3) the interface was so dense that hardly anyone else could get deep enough into it to reap the benefits.
4) in any case it solved a niche problem – which isn’t what Google does.


They’ll celebrate as long as the GOOG dollars trend in the right direction.

When Sun was doing wacky things and growing like crazy it was celebrated. When Sun did wacky things and was not growing, it was frequently dissed by the analysts.


They should continue to have “hobby” projects. Sure, you need to think about the users, buy if you always think from the users’ point of view, you think short.

Jay Cuthrell


I put my take up over at Telecompetitor already but you hit the nail on the head with “impulse to experiment” as being the key factor here. The “impulse to experiment” is what makes it (more) permissible for other companies with lesser R&D resources to gamble on bleeding edge techniques.

Does anyone honestly think that email alone is sustainable for business and collaboration?


Stark Ravin

As an Apple fan, I would also like to celebrate the failure of the Microsoft Kin. Hooray! I hope Microsoft is brave enough to continue engaging in such bold experiments. Now that Google is also showing such bravery, it gives me great hope that they, too, will keep up with such forward-looking risk taking and continue to throw digital spaghetti against the industry wall to see what fails. Heck, we could be up to our neck in delicious Google failure spaghetti. How great would that be? Google, I laud you for shipping one failed product after another. You too could be as great as Microsoft one day.


Disclaimer – I tried Wave and didn’t like it, but…
@Mathew – good post, agree wholeheartedly. Tell Om, though – he tweeted that “Google doesn’t do anything well, except search”.

@Toby – I thought they said they were dropping further development, not removing it entirely.

@PxLated – Interesting philosophical question – depends upon where you end up, I guess. Even development within Corp IT groups has proponents of Agile/Scrum vs. the huge swaths who still favor the Waterfall approach.


Does anyone but geeks/techies like the concept of half-baked ideas, then iterate/update often? I think “normals” like stability instead.

Mathew Ingram

That’s an interesting point, PXLated — but I don’t think non-geeks ever wind up using any of the real beta apps or services like Wave or Buzz, because it never occurs to them or they never come across them for whatever reason. So it probably isn’t an issue in most cases.


It is an issue if you want/need something to be successful and mainstream to carry it forward. Geek/techie network effects only get you so far.

Toby Schumacher

In general, yes it is fine to fail especially on the web. However, Wave was pushed even as late as this spring as a business tool. Imagine if you were a champion of Wave in your corporation and now in 5 months it will disappear. I would imagine any future prospect of both
A) using any new Google technology and
B) career advancement
has just gone out the window this morning. So yes, experimenting then failing is fine for Orkut or trying a new cookie recipe. But for a tool that they were pushing businesses on, no that isn’t acceptable to kill it so soon and leave their biggest proponents out to dry.

Ian Betteridge

I think you’ve missed out the biggest reason for the Schadenfreude: Google massively over-hyped it at Google I/O. I’m still scratching my head about why, as it was pretty clear to almost everyone that Wave was a great technology demo without a clear market. It also didn’t have an obvious route to contributing significantly to Google’s revenues. So why the hype?

Mathew Ingram

Thanks, Ian — although as I said to Leigh, I’m not sure it’s fair to say that it was “massively over-hyped” by Google. I think a lot of that hype came from elsewhere. Thanks for the comment though.

Ian Betteridge

Well, the talk from the stage was, if I remember the quote correctly, about Wave being “what email would be like if it was designed today”. That’s a pretty big statement to make :)


yeah cuz i’m sure Google didn’t have a blogger outreach and PR strategy ;)

They just launch stuff and everyone comes! Google has many products that stay pretty much to the side and don’t get too much press. If something at Google gets hype I find it very hard to believe they didn’t at least light the fires beneath it.


This post seems correct, though I think we’re seeing the difference between Google’s geeky view of success and the market’s business view.


The notion of failing early and failing often is suppose to be about doing smaller betas and getting feedback to make it better. Not sure we are talking about creating behemoths, having huge launches and tonnes of hype only to realize that your product wasn’t relevant or usable.

While I get the sentiment (and the pithy retweets), I’m just not sure it applies here.

Mathew Ingram

Thanks, Leigh — although I’m not sure it’s fair to call Wave a “behemoth” or to say that it had tonnes of hype. It was promoted at I/O but a lot of the hype came from other places, not Google.

Comments are closed.