July 16, 2008

Details of Execution

Sometimes if you do something very difficult, and you do it really well, the end result is that your achievement becomes completely invisible.

Twitter logo

I mentioned a year and a half ago that I like Twitter. That was a little bit less common a position to take back then, but in the months since, tons of people have taken to the little messaging service, so clearly this was no great insight on my part -- it's just a useful, fun service.

But of course, that popularity has not been without its problems. Twitter's gotten a reputation for being unreliable, as a result of its rapid growth. In fact, in many ways, the Fail Whale and its related frustrations has come to define Twitter's brand more than almost anything else.

I'm no expert at these things, but there are a lot of reasons startups fail, and the reasons almost never include the fact that thousands of users clamoring for a service. Indeed, it seems to me that most companies (whether they're tech startups or anything else) fail because of being poorly managed. Put another way, execution is everything.

With that in mind, it's worth pointing out how particularly well-executed Twitter's recent acquisition of Summize has been. I don't know any of the deals of the financial or business arrangements, except that I'm a little disappointed that Twitter isn't maintaining a presence in New York City, instead moving all of the employees to San Francisco. That nitpick aside, the public face of this transition was extremely well executed.

Ev Williams, co-founder and the most public face of Twitter, speaks about the deal at some length in this excellent, candid interview with Techcrunch. (Which site, by the way, may rank as my "most improved" blog of 2008.)

Rumors of the Summize acquisition leaked a few weeks ago, but both companies kept discipline around communications and didn't acknowledge or respond to the conversation. And then, when it came time to announce the deal, the sites had been fully integrated, a lengthy and personable blog post complete with a sketch of some future ideas for integration was posted, consistent branding was in place on the acquired site, and the roadmap for what was going on with employees affected by the acquisition was clearly communicated.

In all, that's a formidable amount of coordination to happen across the country, while business deals are being worked out, and while maintaining secrecy about the fact that it's taking place. And, all of that was done with an eye towards providing a good user experience to their shared customer base.

There are a lot of things to criticize in such deals most of the time, though it seems likely that this will be a successful acquisition, from an outsider's point of view. But what's striking to me is that, as quick as so many are to criticize Twitter (fairly) for technological problems, people haven't been as eager to acknowledge a remarkable discipline and execution on the business side of the company. Frankly, all of those who'd suggested that Twitter should be sold to a larger company seem to have forgotten that almost none of the big companies suggested as acquirers have a history of consistently pulling off this kind of execution. And that's even more true for the smaller innovative companies that they've acquired.

6 Comments

Paying $15m for search functionality on THEIR OWN DATA is excellent execution?

Excellent execution would be figuring out how to add a search feature to their own code, something which takes a few developer-weeks to do (even on Twitter's scale) with Lucene. Paying $15 million because you can't get the engineering resources mustered to do this yourself doesn't really smack of excellent execution to me.

I am a very casual twitter user -- a few tweets per day -- but I have never seen the FailWhale or directly experienced any of the problems that seem to send people off to other services. I hope that twitter stays independent and personable, rather than getting eaten up by a sure-to-be-neglectful overlord.

Twitter proves beyond all doubt that designing scalability at the outset is unnecessary.

Companies fail because they can't find an (paying) audience, not because they can't scale.

I ordinarily like Spolsky but he's dead wrong on this. Simply suggesting that "Twitter could have built its own search for less than $15m" is shallow beyond belief.

I am a casual Twitter user and have only once seen the Fail Whale. It didn’t bother me then and it doesn’t bother me now. For a free service, people (ahem, Mike Arrington) seem to expect more than they should. It’d be a different story if Twitter were a paid service. But it is not, nor will it probably never be.

Given its popularity, Twitter has actually done a great job of staying up under such enormous demand.

I’ve also pretty much given up on TechCrunch. The story last week about Paris Hilton was the straw that broke the camel’s back. I would never consider it “most improved blog of 2008” - not with crap journalism like that. Not to mention that TechCrucnch also seems to be HateTwitterCrunch on days when Mike is in a foul mood. Oh well, there are better tech-oriented blogs out there that are willing to be objective about the companies they are reporting on without letting their personal biases get in the way.

Leave a comment