“Who knows? We’ll probably never know”

Why not just ask them?

Now for some cup cakes and the chronic ….les of Narnia

I think embedding was the killer feature – for the first time people could (easily and at no cost) add video to their own site and the video played inline. If you have a blog, the experience of the video playing inline is just so radically better than linking off to another site.

What I would love to see – YouTube’s traffic broken down by on-site traffic and embedded traffic. How much of that initial growth was from people coming to youtube.com, and how much came from other popular destinations including youtube videos?

Have you considered that Lazy Sunday was merely coincidence and that several other factors, for instance major improvements in site performance or ease of use worldwide that happened at the same time Lazy Sunday was circulated? YouTube is a large organisation with many cogs moving at the same time. It’s hard to accept that we as public observers can speculate accurately what the origin for that growth really was.

The economics literature is full of examples of sites and businesses that are ‘lucky’ – and gives a fairly compelling explanation. Many of you probably already know about network effects, but maybe a rehash is in order. A network good is a good where the utility/benefit to a user depends not just on the good itself but also on the number of other users. Youtube (and similar sites) are classic examples of this, with their reliance on users for content. Other examples are telephone networks (how much would you pay for a phone on a certain standard if noone else had one?), operating systems and many, many other things. Typically, networked goods that fill a similar niche fight each other until a clear winner emerges – if the cost of being wrong is high (I just spent $500 on a betamax video…) then many consumers hold off purchasing altogether. If the cost of being wrong is low (I just signed up for a website noone uses), membership might slowly increase, but use will depend on active members. The result is often that the winner is either (i) lucky or (ii) has the best understanding of how to harness network effects (eg by giving product away cheaply/free early, locking in large numbers of corporate users, etc). Technical differences are often secondary. Bottom line: whatever Youtube’s technical advantages (or not), the reason for their dominance is the network effect.

Ok, guys, so it was network effects. The question is then…how do you start them? That’s what I’m trying to get at…can you design for it?

All history is contingent. All huge successes in software are historical anomalies: right features, right place, right time, right people. That is the fallacy of exemplars. They are based on lightening strikes, not “rules” for success. They ignore the 99.9% of other products/companies who followed the rules and failed. There are no rules in history.

videos that autoplay when the page loads, using flash, and allowing videos to be embedded were the keys to youtube’s success. seems obvious in hindsight, as awesome innovations so often do.

It is likely that great design may have been necessary but not sufficient for YouTube’s success.

If one’s goal is to be a part of creating the next killer web ap, joining an excellent venture capital group that spends significant resources in finding and supporting Mark Zuckerberg type people is probably the best way. VC groups assume a large failure rate as inevitable in finding the next killer ap. They understand that luck and timing are often key ingredients and so you have to sew a lot of seeds in order grow a bean stock that will reach the clouds.

I would imagine that good venture capital groups employ design folks all the time to help support projects, understanding that excellent design is at least necessary, if not sufficient, for big success.

I’d love Josh to write a post about the role of design in creating niche market success. Most web aps/sites are not even trying to be killer aps because their target audience is a niche. My hunch is that the relative importance of design excellence versus luck/timing increases when the target audience is niche rather than universal.

For people in their 30’s and older who need to work and who would like to make a living via some kind of web service or application, it makes about as much rational sense trying to create a killer ap as it does putting money down on the lottery. The risk/return ratio is simply too high.

Understanding what made killer aps succeed is helpful even if, or maybe especially if, we discover that some key ingredients are out of our control.

Great stuff Joshua, Great stuff! Everytime I get to reading your feed, I know i am up for a treat.

I think its meteoric rise was due to a basic social network which combined one of the first easy to use video sharing platforms and was cool for first adopters to use. Plus they used the pirated content as you pointed out. I remember getting IMs that when they were videos all came with a YouTube url and it just got engrained in my brain that YouTube had funny and viral videos. I didn’t use the site as a destination site ie going there in my free time, but I got a ton of clips sent to me from others who found clips on their servers. YouTube got a bit lucky that the big gorillas in the space didn’t hop in faster and didn’t have the coolness factor that YouTube did. Plus it didn’t hurt to ride the popularity of pirated content. There are stats that show SNLs ratings went up after the LS clips zipped around the internet. Good topic and one that we write a lot about over on our blog. Thanks

YouTube wasn’t lucky. Your post explains all the things they got right and after that it was just a matter of time. Lazy Sunday was not the first popular Internet video nor is it the most popular. Lots of sites have hosted popular videos in the past but none became as popular as YouTube on the strength of word of mouth based on the user experience of viewing/sharing a single video.

Hi Dare…I’ve since found some more evidence that they were lucky…at least at the beginning.

it was their name. if it wasn’t that video, it would be the next that would’ve helped them gain traction. so, if you view getting your windshield hit by a rock on the interstate lucky, then riding behind a dump truck increases your chance of luck.

YouTube wasn’t lucky. Your post explains all the things they got right and after that it was just a matter of time. Lazy Sunday was not the first popular Internet video nor is it the most popular. Lots of sites have hosted popular videos in the past but none became as popular as YouTube on the strength of word of mouth based on the user experience of viewing/sharing a single video.

I think its meteoric rise was due to a basic social network which combined one of the first easy to use video sharing platforms and was cool for first adopters to use. Plus they used the pirated content as you pointed out. I remember getting IMs that when they were videos all came with a YouTube url and it just got engrained in my brain that YouTube had funny and viral videos. I didn’t use the site as a destination site ie going there in my free time, but I got a ton of clips sent to me from others who found clips on their servers. YouTube got a bit lucky that the big gorillas in the space didn’t hop in faster and didn’t have the coolness factor that YouTube did. Plus it didn’t hurt to ride the popularity of pirated content. There are stats that show SNLs ratings went up after the LS clips zipped around the internet. Good topic and one that we write a lot about over on our blog. Thanks
For people in their 30’s and older who need to work and who would like to make a living via some kind of web service or application, it makes about as much rational sense trying to create a killer ap as it does putting money down on the lottery. The risk/return ratio is simply too high.

Understanding what made killer aps succeed is helpful even if, or maybe especially if, we discover that some key ingredients are out of our control.

I think Shai is right. The design was necessary but not sufficient. The graph shows them taking off in early 2006, but they only released their beta version a few months before that. Their wheels were greased with good design and it was just a matter time before some great viral video came along and propelled them. Also, I remember when I first heard about YouTube (in my then GF’s apartment in Harlem) in early February 2006. I had also recently played around with Google Video. And Google Video had restrictions on the format and size of the vids you wanted to upload. Right then I thought, “well I’m not using Google anymore. I’ll use this YouTube thing.”