Ask HN: Any resources for solving the chicken-egg problem?
I am preparing our startup launch in the next few months and I would appreciate if someone has any resources about this problem (how to get users to register because your site needs users to function). It is based on the principles of uber and airbnb.A two-sided community/marketplace. How did these sites start? With promises? Local campaigns? Global launch through an app? This is also a question in YC's application form.
13 comments
You might want to give this a read: http://paulgraham.com/ds.html ("Do Things That Don't Scale").
It's the best piece I've ever read dealing with the question of attracting users to an early-stage startup. The gist of it is this: your first 100 users will be completely different from your last 10,000. The first 100 have to be courted - you're going to have to bend over backwards to make them happy. If that means customer support emails from the founder, so what? Those first 100 users, the ones who are overwhelmed by your support, will be the most valuable customers you'll get, because once they're through the door, network effects can really start to kick in.
The apocryphal story I've heard (but never verified) is of how PayPal broke out in the eBay marketplace. Instead of trying to convert customers with advertising and the like, they pretended to be customers on eBay, offering to buy inexpensive items from popular sellers before asking if the sellers accepted PayPal. This got the first few sellers through the door, and once they started using it the network effects propagated through eBay and beyond, turning PayPal into the giant it is today.
Hope this helps. Good luck!
It's the best piece I've ever read dealing with the question of attracting users to an early-stage startup. The gist of it is this: your first 100 users will be completely different from your last 10,000. The first 100 have to be courted - you're going to have to bend over backwards to make them happy. If that means customer support emails from the founder, so what? Those first 100 users, the ones who are overwhelmed by your support, will be the most valuable customers you'll get, because once they're through the door, network effects can really start to kick in.
The apocryphal story I've heard (but never verified) is of how PayPal broke out in the eBay marketplace. Instead of trying to convert customers with advertising and the like, they pretended to be customers on eBay, offering to buy inexpensive items from popular sellers before asking if the sellers accepted PayPal. This got the first few sellers through the door, and once they started using it the network effects propagated through eBay and beyond, turning PayPal into the giant it is today.
Hope this helps. Good luck!
It's nice to build up a roster of pre-signups before you launch. It's super easy to throw up a launchrock.com page on your site. Collect emails while you're building the product, then when you launch just send a blast to that group. Should get you the initial inertia you need. It's also worth trying to get your thing listed on sites like betalist and producthunt.
But in general, I really wouldn't worry about it. Early adopters LIKE to be first. So while a normal user would be turned-off by an empty marketplace, your early-adopters won't necessarily see it that way. They'll help you build out content and fill up the marketplace, so that normal users find it attractive down the line.
But in general, I really wouldn't worry about it. Early adopters LIKE to be first. So while a normal user would be turned-off by an empty marketplace, your early-adopters won't necessarily see it that way. They'll help you build out content and fill up the marketplace, so that normal users find it attractive down the line.
I honestly don't want to be mean here. But I worry for you that a startup with such an obstacle should have had a plan of attack from the beginning, and even then would have been a long shot. Is rapid early viral growth your only hope now?
I didn't launch yet. we are "in the beggining" and certainly not hopeless :)
Is your community a one-sided community, two-sided community, or marketplace?
If it is the second or third, you ought to read HBS Professor Andrei Hagiu's case and article about this dynamic: http://www.hbs.edu/faculty/Pages/item.aspx?num=24044
Platform Thinking also has a pretty comprehensive article on this: http://platformed.info/seeding-two-sided-businesses-strategy...
My two cents here is that the best place to start in disrupting the chicken and egg problem is to reduce the number of users you need to get off the ground. Scope to as small a group as possible to demonstrate that the app provides the value you are hoping to provide, and then grow from there. If you're making an app for teenagers, start in one high school. If it works, grow to two, then four, then eight, and so on. You can scope geographically or by affiliation or in any of a number of other ways.
Good luck.
If it is the second or third, you ought to read HBS Professor Andrei Hagiu's case and article about this dynamic: http://www.hbs.edu/faculty/Pages/item.aspx?num=24044
Platform Thinking also has a pretty comprehensive article on this: http://platformed.info/seeding-two-sided-businesses-strategy...
My two cents here is that the best place to start in disrupting the chicken and egg problem is to reduce the number of users you need to get off the ground. Scope to as small a group as possible to demonstrate that the app provides the value you are hoping to provide, and then grow from there. If you're making an app for teenagers, start in one high school. If it works, grow to two, then four, then eight, and so on. You can scope geographically or by affiliation or in any of a number of other ways.
Good luck.
these are good articles, I was searching for that kind. it's a two-sided community/marketplace a combination of the two.
You mean website that need user generated content? Even a big sites like reddit/digg/twitter started with just the founders and friends being very active, sometimes creating multiple accounts (I'm just reporting what I've read, not suggesting you should to that).
Classifieds sites are often seeded with crawled data or offline data or manual data entry.
Airbnb and uber I think started with approaching sellers and drivers directly. That means cold calling, offline ads or meeting them in person. And probably it started months before the launch and included some kind of incentive like "no listing fees for you in the first year". It's like a waiting list: people sign up, but they can't see the product yet. Only after there are 100s of users you show them the product.
Classifieds sites are often seeded with crawled data or offline data or manual data entry.
Airbnb and uber I think started with approaching sellers and drivers directly. That means cold calling, offline ads or meeting them in person. And probably it started months before the launch and included some kind of incentive like "no listing fees for you in the first year". It's like a waiting list: people sign up, but they can't see the product yet. Only after there are 100s of users you show them the product.
One solution is lots of money so that you can buy your way into both marketplaces quickly.
A better solution as jchung pointed out is to narrow your scope until you have enough resources to create a great market. For example:
USA -> California -> Bay Area -> Stanford -> One Dorm
A better solution as jchung pointed out is to narrow your scope until you have enough resources to create a great market. For example:
USA -> California -> Bay Area -> Stanford -> One Dorm
This is the 'get a small number of users to love your service' approach that I have seen mentioned alot in solving the chicken and egg. It is how Facebook got going.
how to do that if my service doesn't involve professionals? uber may had to spend money to convert taxi drivers but airbnb used ordinary people. you mean through advertising? that too needs people to be already present in the platform..
Are you asking how airbnb could have spent money to attract users? For example, they could have spent money to provide a discount (charge the guest $100, pay the host $200, both are more likely to join than at a market rate of $150). Also, they could have acquired hosts by offering a guarantee, such as paying for 3 nights a month if they don't have at least 3 nights worth of guests. There are also more general techniques like advertisements and salespeople and such.
There is a way that has worked for established companies like Ariba, Quora, etc. It has also worked for me to some extent.
(I also got a response to cold emails that I sent to ariba and few other founders).
1. Fake one side of it : This should be applicable to most marketplaces but not all. For ex, you can fake the users who answer questions on Quora.
2. Fake both sides : This is a bit extreme but definitely a practice used. For ex, fake users asking questions and fake users answering on quora. This basically starts to create momentum validating it for new users.
3. Content : Content is a new way to solve this. This is something I have tried first hand. Create a blog on the same domain. Start creating content that explains your marketplace and starts to attract both sides of your marketplace. You can also strategize so that you start attracting the heavier side of users first( ex : buyer/seller case. If you have buyers you will end up having sellers soon).
One really important note is to see the cause and effect of your marketplace. Which side of marketplace will be easier to build if you have the other built.
Hope it helps, All the best !!
1. Fake one side of it : This should be applicable to most marketplaces but not all. For ex, you can fake the users who answer questions on Quora.
2. Fake both sides : This is a bit extreme but definitely a practice used. For ex, fake users asking questions and fake users answering on quora. This basically starts to create momentum validating it for new users.
3. Content : Content is a new way to solve this. This is something I have tried first hand. Create a blog on the same domain. Start creating content that explains your marketplace and starts to attract both sides of your marketplace. You can also strategize so that you start attracting the heavier side of users first( ex : buyer/seller case. If you have buyers you will end up having sellers soon).
One really important note is to see the cause and effect of your marketplace. Which side of marketplace will be easier to build if you have the other built.
Hope it helps, All the best !!
Paths to success in this realm normally involve a lot of luck and moxy. For some insight, see the Airbnb interview on Mixergy
http://mixergy.com/interviews/airbnb-chesky-gebbia/
http://mixergy.com/interviews/airbnb-chesky-gebbia/