I spend quite a bit of time talking about leading-edge Web stuff to mainstream Enterprise types. I have a well-polished explanation for the rise of PHP and Rails and so on: Time To Market. Here’s the sound-bite: “If you and I have the same good idea for a community-based Web site on the same day, and mine is on the air in five months and yours in eight, then you’re dead. And it doesn’t matter if yours is better, because the community has gathered.” Well, Twitter would be the canonical example. They went with Rails because it let them build fast; and they built fast. They suffered terrible pain for months trying to take Rails places it’d never been before; but they fought through it and they’re in a very good place. Smart people tell me that Pownce and Jaiku are slicker and better but who cares? Apparently 140 characters, distributed appropriately, gives you what you need.
Contributions
Comment feed for ongoing:
From: Ben (Mar 14 2008, at 06:03)
Yes and no. Case in point for no: Google.
Time to market is important, but it's not a death sentence for late-comers.
[link]
From: Noah Witherspoon (Mar 14 2008, at 08:44)
Myspace / Facebook?
[link]
From: Rory (Mar 14 2008, at 08:44)
Another counterexample would be facebook, which leapfrogged myspace (and completely obliterated earlier services such as Friendster).
[link]
From: Greg Phillips (Mar 14 2008, at 08:50)
Ah, but Ben, Google (search) isn't a "community-based Web site", it's an algorithm-based service.
There are other counter-examples, like the whole Friendster, Orkut, Myspace, Facebook trajectory, but I don't think they completely kill Tim's argument.
[link]
From: Jeffrey Fredrick (Mar 14 2008, at 08:57)
WRT Google that is an example of the front runners throwing away their advantage. AltaVista had the community, it was theirs to lose, and they lost it because they didn't solve the problem well enough. (Clearly if they had come to the market _after_ Google we would never have heard of them.)
Consider MapQuest vs. Google Maps. MapQuest was there first, I think Google Maps is far better, but MapQuest remains widely used because they were there first.
[link]
From: David (Mar 14 2008, at 08:57)
Tell it to Friendster.
[link]
From: Roger Wilco (Mar 14 2008, at 08:59)
Another exception/addition: MySpace overtaking Friendster.
For community-based sites, I'd say the most important thing is the community. Pownce may be better than Twitter but the community is, more-or-less, identical. So why switch? Jay-Z didn't have a Friendster page, but he has a MySpace one. I know which site I'd rather hang out at.
[link]
From: mummey (Mar 14 2008, at 09:08)
Counterpoint: Friendster -> MySpace -> Facebook
Sometimes its the crowd itself that makes kills the 5-month site and causes its users to flock to the 8-month site.
[link]
From: Adrian Madrid (Mar 14 2008, at 09:09)
I guess you can say that Google also proves the point: they had to make it so much better than the competition to win. And they did. I guess the point is that arriving first gives you the community and raises the bar quite high for anybody coming second.
[link]
From: Nathan (Mar 14 2008, at 09:09)
" for a community-based Web site"
No, still mostly no. Google is not a community-based website. The idea of "community has gathered" is central to his point but doesn't apply to Google.
[link]
From: Anson (Mar 14 2008, at 09:13)
Also, the real competition for Twitter is really the status field in Facebook and Myspace.
At least for the non twitter-as-im-client crowd.
[link]
From: Mihira Jayasekera (Mar 14 2008, at 09:16)
@Ben: If by "Google" you mean just the search engine, that's not really a great counterexample. Tim Bray's quote was specifically about community-based sites, which Google (the search engine) is not. Google-owned YouTube might actually be a good supporting example of Bray's point -- not necessarily the best execution of community-based video sharing, but the first.
[link]
From: John (Mar 14 2008, at 09:16)
Ditto what Ben said with MySpace v. Facebook.
[link]
From: Andrew (Mar 14 2008, at 09:18)
I wouldn't consider Google a valid counter-point, as a search engine is not exactly community based.
[link]
From: Bill C (Mar 14 2008, at 09:19)
I think Twitter serves as a good example of where something simple benefits from time to market. Search is arguably more complicated, so getting it right is more important. I suspect that means complexity can overcome first mover advantage if well executed.
I would add that both benefited from highly visible commentary and more superfically, more sensible or descriptive names.
In other fields, I wonder if the Tripit, Dopplr, Vibe Agent, other services market will work similarly to Twitter or if there's room for 3-4 of these programs.
[link]
From: Mike Purvis (Mar 14 2008, at 09:20)
Search engines don't have network effects, though. Who wants to use some other service if all the cool people worth following are on Twitter?
[link]
From: Dominic Mauro (Mar 14 2008, at 09:22)
Ben, that's an excellent example, but hasn't the environment of the internet changed since the rise of Google? Further, Google isn't so much a social network as an individual web-app, I would argue.
[link]
From: julian (Mar 14 2008, at 09:23)
@Ben: Ah, but that's why Tim has the "community-based" qualifier in there. I'd argue that your search engine is not community-based.
One could make a case for something like Friendster vs. Facebook, but there are always other forces at play.
[link]
From: Assaf (Mar 14 2008, at 09:43)
You have to be first, but also be good enough to keep that advantage, or you'll be bleeding community.
So one thing that often gets ignored is being able to evolve at a reasonable rate. You need that flexibility.
I've seen my share of tools that focus on time-to-market, which we don't hear about. They make it really easy to create instant legacy code that no one can change.
The other thing is variety. We remember the winners, but who remembers the losers? Look at open source, we've got some game-changing software out there, like Apache or Firefox. And a million no-go on SourceForge. So which is it? Both.
Chances of hitting home run are bigger if you get more than one round in the game.
[link]
From: Christian (Mar 14 2008, at 09:46)
The nice people over at Facebook would probably disagree with you.
[link]
From: Chris Patterson (Mar 14 2008, at 09:52)
Google's value to users isn't the community, though.
I think another way to reframe this would be "community-based sites have a magnified first-mover advantage".
I don't think it's necessarily an insurmountable advantage, but if the first site is "good enough" for users, the value added by a vibrant community can become a powerful positive-feedback loop, making it progressively harder to unseat the incumbent.
eBay and CraigsList are great examples of "good enough" community-based sites. Neither has a huge edge in terms of technology or interface, but both are pretty dominant in their target space.
[link]
From: Kevin Ballard (Mar 14 2008, at 09:54)
Ben, Google isn't a social website. It doesn't depend upon a community to work.
[link]
From: Stefan Hayden (Mar 14 2008, at 10:00)
google is a good example.
readburner might also be a good example. They were out first but did not have the time to devote as RSSmeme did.
I think the trick might be that if two people come out with the same idea then the second on will fail.
But if the second one solves a fundamental problem with the previous one then it has a good chance.
Google solved a fundamental problem of yahoo, which I would say search and speed.
RSSmeme solved a fundamental problem of readburner, which I would say is development speed.
Does Pownce solve a fundamental problem of twitter? I doubt it. It has a few bells and whistles more but that's not enough to attract market share.
[link]
From: brookr (Mar 14 2008, at 10:15)
Holy crap, does anyone read the comments before posting?
20 people just made the same 3 points (only one of which is valid).
Social networking sites are a subset of "community-based" sites... and have their own set of rules. people move on to a new social network site when they "friend" too many people who they aren't friends with, and don't feel good about "unfriending" them. Twitter provides a very easy way to block someone without offending them, so you don't have to "find the next one" when things get awkward. With social networks, a clean start is really enticing to people.
Something new will replace facebook in a year.
[link]
From: Assaf (Mar 14 2008, at 10:17)
In my circle, Google use spread not from ads on TV (contrary to other search engines), but word of mouth, so the community played a big part in getting people to switch over.
Back then Google wasn't just a better AltaVista, it was a search engine that actually returned relevant results. It was as different as Tumblr and Twitter (both do microblogging). Time to market still applies.
If someone else did page rank a few months before, would Google exist?
[link]
From: Johan (Mar 14 2008, at 10:44)
You only have to convince ten or so of your closest friends, those who you spend the majority of your time talking to, to switch, not the whole network. Agreed, if you look at a network as a whole you might say: All of those people wont just switch one day! Switching happens on the margin.
[link]
From: Brian Dame (Mar 14 2008, at 10:49)
This is assuming that the entire potential community will gather in that span between releases. I don't think three months is a long enough time for any company to gain such a significant foothold that competition is insignificant.
[link]
From: duffy brook (Mar 14 2008, at 10:55)
One might argue that the Friendster / MySpace / Facebook (F/M/F) comparison depends to some extent on the "community" in question and its purpose, as well as the features. In other words I think the above is an apple / orange / pear comparison to some extent.
If you have two webapps that address the same type of communities with similar features, then the first to market should dominate.
My personal experience with F/M/F is:
Friendster: too ad-heavy, too open to non-friends, my friends either didn't join or didn't find use for it, I couldn't establish a community and stopped using it after a couple months.
MySpace: too ugly, too teen-centric, bad press about predators, okay if you're promoting your band on the cheap, more "me too" than meaningful. Community = high schoolers and band fans
Facebook: clean interface, respects privacy, features and customization options are more sophisticated, lots of ways to connect with friends. Community can exist on many levels (schools, employers, hobbies, interests, music, and more), or between just a few...
For my money (ha), the pear tastes best regardless of which was first.
[link]
From: Emil Sit (Mar 14 2008, at 11:56)
This is a restatement of the classic "Worse is better argument.
[link]
From: Carlos Pero (Mar 14 2008, at 12:06)
The example of 5 months vs. 8 months holds true if two people have the same idea on the same day.
On the other hand, the Web is not going away, and there is always opportunity to build a better product. For example, nothing is guaranteeing that the people on Twitter won't gravitate towards a new product that will better meet their needs. (Some people are already ready to leave Facebook and form a new community if something new pops up.)
[link]
From: Trout (Mar 14 2008, at 12:48)
Do you remember Friendster?
The community had already gathered.
You could say the same about other services pre-dating Friendster (Firefly was it?) They were certainly first. When a better service came along (one not as slow, and allowed some things the users were calling for, but that Friendster had rejected) people finally had it, reached a tipping point, and switched en masse.
As MySpace becomes slower by the day and very confused and harder to use visually, and Facebook continues with it's mildly arrogant stance towards user privacy and choice -- don't think it can't happen again.
[link]
From: Ben (Mar 14 2008, at 12:48)
Ack, sorry - should have read more carefully before commenting.
Honestly, however, I don't think the first mover argument is limited to social sites. In most markets, the first mover has some advantage, but given a better-enough competitor (by actual or perceived quality) that can be overcome.
As a non-web example - DVDs overcame videotapes (and laserdiscs) despite the financial pain incurred by transitioners (they had to be entirely new machines).
[link]
From: Jemaleddin (Mar 14 2008, at 12:50)
Every photo site vs. Flickr. Nothing new, just better thought out.
[link]
From: Chris Pirillo (Mar 14 2008, at 14:19)
Pownce? What's Pownce?
[link]
From: Adam Rice (Mar 14 2008, at 16:29)
Twitter benefits from a lot more than time to market:
1. Easy API. Lots of developers support it in lots of ways.
2. Easy to use. Some of the Twitter-likes aren't quite as simple.
3. Fun.
4. Flexible (see above). It's a chat/SMS/nano-blog thing.
Right now Twitter has massive buy-in from people like, well, Tim Bray. If Twitter did something to piss off its nerderati users, or simply wound up having unacceptable outages (and it has been kinda flaky), and if there were an otherwise equivalent service that happened to be later to market, its current user base would leave in droves.
Some first-of-their kind socially oriented services are succeeding, like Twitter and Delicious. Not all are. And many have specific constituencies, mostly by accident. Harry Potter fans are disproportionately represented on Shadows, Brazilians on Orkut, and Ron Paul supporters on Digg, apparently—these accidental constituencies are more interesting. Twitters is nerds. So just because Tim sees everyone he knows on Twitter doesn't mean everyone is there.
[link]
From: Mark (Mar 15 2008, at 05:22)
This is tech-centric (as I find many "laws" of the internet to be).
In many niches (including the one my company operates in) there are several successful community sites that are very similar, but which have a slightly different style and vibe. In fact, I would say that there is a point at which a community gets too big and drives many participants to smaller communities.
(I'd add that our niche is aimed at ordinary women, mostly thirty-/forty-something, a group less subject to internet herding than twenty-something male tech geeks and more tuned in to the subtleties of a community's ambience.)
I think your point makes sense in market-making apps like eBay in the U.S. or Yahoo! Auctions in Japan.
[link]
From: Greg Paulhus (Mar 15 2008, at 05:51)
Re: "If you and I have the same good idea for a community-based Web site on the same day, and mine is on the air in five months and yours in eight, then you’re dead. And it doesn’t matter if yours is better, because the community has gathered."
This isn't strictly true, for a few reasons. One, the site that launches first has to be 'good enough', it can't be junk. Two, the site that launches second has to be mostly similar in all features and functionality. Three, there is in fact a degree of 'better' which will shift the audience, and others have already pointed out many examples. Four, the site that launches first has to find the market effectively, and that doesn't always happen.
To sum up, I do agree that it's smart to get to market quickly, but I do not agree that there can never be a better Facebook, Twitter, Flickr, etc (and history has shown us this is true).
[link]
From: Stephen Downes (Mar 15 2008, at 10:48)
If I have an idea at the same time as Tim Bray, and my idea is online in five months, and Tim Bray's idea is online in eight months, then I'm dead, because I can't build the kind of community around an idea in three months that Tim Bray can build in three minutes.
Time to market is important. It is, however, trumped by fame, connections, and corporate support.
[link]
From: Gerry (Mar 27 2008, at 15:44)
This is something Al Ries and Jack Trout wrote about in their 1993 book "The 22 Immutable Laws of Marketing," so the idea's been around for a long time. It's significant enough to be the first of their 22 laws. Law #1: The Law of Leadership - It's better to be first than it is to be better.
I have to say that the comments here strike me as an example of why age and experience are sometimes pretty darn useful. Lots of folks here seems to be debating this as if it's a new concept.
[link]