Especially at Google. Everything has to be ready for a billion customers on day one. While this works great for a lot of products, I contend this is one of the big reasons why they have a lot of problems with social.
The great social networks find a band of early adopters who get to know each other on that system, and they all figure out what this thing is, while the dev team responds to what they do, just as fast. At least at the start, it's like a little coffeehouse that a clique discovers and makes their own, and starts holding events there.
When Google launches a social product it's like going to the opening of a mall. The place is glitzy, but you're lost in the crowds.
Out of pure curiosity, why couldn't Google just invite a few thousand people as a kernel to try things out? Wouldn't be that hard to find a couple of social supernodes and their friends, beween GMail and AdWords...
My information is way out of date, but I believe most Google products are still tested that way. There's a long Googler-only incubation period for everything. And Robert Scoble was probably on Google+ months before you knew about it.
The people at Google aren't stupid. They know they need to iterate with customers. But they're iterating towards a very specific event, the day when they turn it on for a billion people in 10 languages.
I'm saying that very wide launch makes it kind of sterile. And the drive towards scalability-first probably prevents them from exploring quirkier ideas.
This is exactly what they did with Wave and is one of the reasons why it failed; you always end up with some group of 8 people that can use the system and 2 people that can't making the entire product useless for collaboration.
True. I think the problem is that, when blinded by the "we're going to be huuuuuuuge" mindset, it's easy to forget that shit needs to work at the small scale first.
