Hacker News new | past | comments | ask | show | jobs | submit login
Ask PG: Should network effects be a turnoff?
44 points by adamzerner on Aug 4, 2013 | hide | past | web | favorite | 21 comments
In How to Start a Startup, you say:

"I can think of several heuristics for generating ideas for startups, but most reduce to this: look at something people are trying to do, and figure out how to do it in a way that doesn't suck."

and then go on to give the example of making a better dating site. But what about network effects? That's basically the name of the game with dating sites.

1) Should small startup founders avoid ideas that involve network effects?

2) If not, what sort of heuristic would you propose for a small founder to overcome network effects?

Network effects: http://en.wikipedia.org/wiki/Network_effect


"8. Dating. Current dating sites are not the last word. Better ones will appear. But anyone who wants to start a dating startup has to answer two questions: in addition to the usual question about how you're going to approach dating differently, you have to answer the even more important question of how to overcome the huge chicken and egg problem every dating site faces. A site like Reddit is interesting when there are only 20 users. But no one wants to use a dating site with only 20 users—which of course becomes a self-perpetuating problem. So if you want to do a dating startup, don't focus on the novel take on dating that you're going to offer. That's the easy half. Focus on novel ways to get around the chicken and egg problem."

To overcome network effects, such as in dating sites, the following can be used:

- Leverage an existing network (ie: Facebook).

- Partner with an existing network (ie: target niche users of a dating site)

- Do not make it look/feel like a dating site, but gradually grow into it, or keep dating as an aside functionality to finding or hanging out with like-minded people (much like a bar).

- Seed the dating site with fake or paid users (seedy, but practical).

- Give free or low-threshold access to either the chickens or the eggs. A dating site that is free for women or offers different functionality.

- Hustle your early users. Make friends and family use it.

- Making it useful for online marketeers, exclusive invite only, or the "place to be" by attracting famous people.

Hustling your early users is a great idea - I'd put this at the top of the list.

Planning to grow into something else is dangerous, though. To paraphrase Peter Thiel, you should be wary of building a startup that's a two-stage rocket. If the first stage works, you can and should stick with it; if it doesn't, you'll never make it to stage two.

Someone good enough to execute a two stage startup on purpose (Bezos, Musk) are not relying on luck as compared to the rest of SV startup lottery.

I think many of the young and arrogant "kid CEOs" would be better off in the long run by overcoming their ego, go and hustle the smartest mature CEO they can find into hiring them. Defer taking your own risks until you are actually in control of them.

It's interesting that Reddit was used as an example of a site that doesn't have a chicken and egg problem. Reddit admittedly seeded the beginning with fake users.

What about "launching"? Once you have several thousand emails collected, start setting up dates all at once, avoiding the chicken and egg problem.

I would also add to that list

- target a very narrow, undeserved market and when successful expand beyond the core market (ie: Craigslist and its local focus)

So long as you have smart ways of targeting that narrow, underserved market at a reasonable price.

The essence of PG says is that you have to bring down critical mass.

For that reason, a founder has to make the product/app so that it's even useful only with several hundred users.

If your product is not useful with only a several hundred users, you built the wrong features. --> Try again.

Tremendous answer. Seems frequently overlooked that a product should work great for the first 100 as well as it does the first million.

If you get 50 people to come back everyday to your site for a week then you probably have something special.

This answer is right on spot, short and clear. It's the best answer. I'll add that by critical mass we use the analogy of a nuclear reaction fuel where a self sustained chain reaction can take place.

I don't think so. It's not true that the set of good companies only includes those that would be good with only a couple hundred users. There exist a lot of companies that need lots of users to be good. That need network effects.

I've been thinking about this too, but which successful startups that underlie network effects worked only with several thousand users? There are market places and "solomo" apps as far as I can see, market places should work with a few hundred users (for that reason pick a niche), however nobody has managed to make solomo work yet, because all of the solomo apps only work with several thousand users in an area and it's just impossible to get to that number of users in a sustainable way.

Instagram, Snapchat don't underlie network effects in the beginning really as they have circumvented the chicken and egg problem by having a great single player mode with filters in Instagram and on Snapchat you only need one friend to send funny pictures with.

The bell phone network strated with less than 100 users in the beginning, as well as radios and tvs. These are all examples of very stong network effect dependent businesses.

All of these businesses had at some point less than 100 users. At some point the self sustained chain reaction took place. This is what makes network effect dependent business attractive and also because one can easily grab and lock the full market keeping competitors away.

The key point is how to reach the level where the chain reaction starts. For that you need tactics which increase the value with the few people available.

For a phone company trying to sell phone where there is no such thing, offer a phone line to a relative with each phone line bought. Offer phone lines to service providers like doctors, hair cutters, garage, etc. This increase the value of buying a phone line.

Facebook, did just that, without knowing it, by starting with Harvard students. There was high value to be part of it and the i initial people made it very attractive. This is lowering the critical mass. No one wanted to look at MySpace.

There is far more to business than SV start ups funded by VC.

I think network effects should be viewed as a tool for growth, but not a requirement for early adoption.

As long as you are doing something in a new, better way for some number of people, you will have early evangelists who love it. Then network effects will help you grow quickly and become valuable to a wider population.

I think Airbnb is a good example: they had to fight hard for their early users, but the product had enough inherent value that network effects served only as a value multiplier.

If you think there are network effects, the odds are that they are much smaller than you think at scale. As http://www.dtc.umn.edu/~odlyzko/doc/metcalfe.pdf explains, usually they are of strength n log(n). For small n you can address that with a degree of personal attention that larger companies can't handle. For larger n, you have to be better than your competition for your use case, but not massively better.

If you have true quadratic network effects - auctions would be a good example - you probably need to find a different startup idea.

The network effect is just one barrier that helps make your company more defensible from competitors. It's not a magic bullet though. Just ask MySpace or AOL.

You should follow PG's advice to make something people want, but also try to think of barriers you can construct along the way to protect the value you create. Keep in mind that most small businesses don't have a strong network effect, yet many are very profitable and sustainable.

I don't understand what that small paragraph from one of pg's essays has to do with network effects.

It seems to imply that you could just find things that suck, make them better, and people will come.

OP seems to be reading "figure out how to do it in a way that doesn't suck" as "imagine a way that wouldn't suck."

'Figure how to do it' is nice and specific: for a physical product, figuring out how to do it so it doesn't suck includes manufacturing plans. For things that have a network effect, figuring out how to do it so it doesn't suck includes having some marketing plan that lets you get it off the ground. e.g.: big launch; minimally seeded community (e.g. for dating, find every world war I reenactment enthusiast to join - maybe htere are only a few thousand of them - and iterate on them to begin with); fake having the network effect already; or simply piggy-back on someone else's network through some kind of arrangement.

"Figuring out how to do it" includes, well, figuring out how to do it.

It's not enough to imagine how it might be better. You have to imagine and find a path to it that you can actually do.

If you can't do it, pick a different idea! It's called a barrier to entry for a reason; you're not going to figure out how to create a solar-powered electric flying segway if you only have an arduino and a few scrap parts to do it with. If you can't think of a plan, then pick something you can think of a plan for.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact