Besides a killer algorithm and brand-name recognition, Google's greatest strength is its speed at releasing new products. We get to play with new, cool and ever-improving tools for free. Recently, though, we've seen that being unwitting lab subjects can kind of stink.
"Labs" we love: Gmail & Google Wave
Google has practically reinvented how web applications and software are developed with their overriding love of "Beta", a term that used to indicate a product that was functional and mostly stable, but not quite something the company would stand fully behind or "publish", whatever that means now.
Gmail, arguably Google's greatest popular success (meaning not flush with advertising cash) outside of web search, started out as an invitation-only service on, famously, 1 April 2004, and rolled out to the public in February 2007. Even after those nearly three years of controlled growth, Gmail stayed "in beta" for two more years. Google ultimately admitted the "beta" tag had little, if anything, to do with where the product stood in the perceptions of its developers or users. While Gmail was still "in beta", the service also introduced a Labs section, where the type of people who explore their apps' settings could turn on cool features Google wasn't quite sure everyone would use, or like the looks of, but that the development team felt were pretty neat additions.
That strategy works. The sense of being in on the ground floor of a smart company's ever-improving product made evangelists out of the early adopters, and Google was spared the anguish of having to hear the masses' feedback on every keen idea their developers had — if you didn't like something, just don't turn it on. More than that, though, the constant tweaking didn't hurt the product's core offering. No matter what, your email arrives, a surprising amount of spam is blocked, and the search, filter and label functionality are rock solid. If you want more, you could cherry-pick it from the Labs section or, as many of us did early on, find browser add-ons and user scripts to fill your needs.
For a lot of Google's other products, this model works just as well. Maps gets you from A to B, but it has its own Labs to play with. New features on Google's search and results pages go out first to a semi-random group of users, and some of the most bold changes that get finalised — SearchWiki, personalised searches — can still be turned off.
Let's not forget the news and hype value of roll-outs, regular iteration and open secrets. This site is as guilty as any of giving Google a lot of attention for their little changes, because they're constantly occurring and widely covered — a chicken and egg dilemma we're well aware of. Even when Google wants to keep something "secret", it's at best half-hearted. They denied their development of a directly-approved-and-sold mobile phone with misdirecting quotes, and asked thousands of employees at one of the world's most connected companies not to talk, tweet or blog about the brand new phone they all received as a holiday bonus. Even if the Nexus One hasn't broken any sales records, it certainly grabbed attention and over-wrought headlines.
For Google Wave, a product that might actually be a potential revolution, a slow roll-out with eager test subjects makes sense. Developers and power users can report the problems and suggest solutions, see how their own dreams might fit in, and figure out real-world use cases. Aside from some initial "What's it do?" scepticism, the expectations of both Google and its users seem fairly compatible in the great Wave sandbox, and everybody's free to walk away with no harm done.
"Labs" that leave us chafed
How does Google's mindset of release first, iterate later fall down? When it leaves early adopters and paying customers feeling left behind and little cared for. In three cases, you can see how Google applied a software solution to a product problem, and left many with a sour impression.
Android mobile phones: The idea of Android itself was appealing from the get-go: A free, open-source platform that any phone manufacturer can build for, with an application store that's hardly restricted at all. We, the geeky public, wanted that phone.
Unfortunately (at least down under) we often didn't get that phone. While there's a relatively healthy selection of Android phones out there (particularly from HTC), the fact that Android is open source hasn't led to a glut of choices, with many of the US models staying resolutely on that side of the Pacific. Indeed, in some cases phones have been announced only to be pulled at the last minute. Google itself is guilty of this: the Nexus One is yet to have an official Australian release plan announced.
One issue with this, demonstrated amply by the US experience, is a lack of consistency. you were writing a user's guide to Android, or even just trying to explain its features to a friend, you'll sound ridiculous. "How do I check out the photos I've just taken?" "Well, you click the little thumbnail in the upper-right corner, unless you've got a Hero from Sprint or a CLIQ from T-Mobile." "Can I adjust the white balance or zoom level in the camera?" "Sure, if you've got a Droid from Verizon or a Nexus One. Otherwise, you'll have to wait." Elsewhere in the smartphone market, owners of the very first iPhone are running basically the same OS as someone who buys one today, minus features (3G, true GPS, video recording) that have been very clearly spelled out in each subsequent hardware upgrade.
In an ideal world, Google and its hardware partners could sell any phone and have it work on any network, and OS updates would come from Google, direct to all phones. In the meantime, Google's experimenting with new OS builds, seeing which carriers and business models work best, and leaving its end consumer confused as to what they're buying into.
Google Buzz: The blow-by-blow of what Google did and didn't do right by its users when it suddenly dropped a new Buzz social network in their Gmail inbox is intricate, but neatly summarised by the BBC. Gmail and Buzz manager Todd Jackson suggests the product needed more testing before getting such wide implementation, and that seems true, in a way: it needed more real users testing it. Had Google stuck to a gradual release to enthusiasts and early adopters, Buzz might have worked out its kinks more quietly, and Google wouldn't be stuck with a recurring headline about Buzz invading users' privacy.
Instead, Google made us all "Labs" testers. The fixes and improvements came fast and furious, but the company that famously set its primary goal as "Don't be evil" decided, in this one instance, that quickly obtaining a social network with tens of millions of users was more important than giving the users of one of its second-most important products a gradual learning curve. That's troublesome. Buzz is, again, not a mandatory or even productive product, and can be entirely turned off — but the clear path to doing so required unwitting test subjects to ask for it.
The point: Openness is always better
Google's a lot more "open", in terms of communications, disclosure and software/API standards than most tech firms. In certain areas, though, they seem to take an old-school approach to releasing new products, without enough thought into where sudden launches and follow-up improvements will leave their enthusiastic early users.
As one of the users who gets almost everything I do on the web from Google for free, minus the cost of our eyeball allegiance, I'm hardly in a position to say Google's somehow cheating me, or not trying hard to provide exciting services very quickly. In doing so, though, I hope they'll give more consideration to what they can offer as a mostly harmless beta test, and when they should wait to ship as an actual product — as Old Economy as that may seem.
Every now and then, we like to go on grumpy, long-winded, opinionated rants. We're far from the definitive voice and your feelings may differ, so feel free to air your thoughts in the comments.