Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: How do sites like Reddit, HackerNews, etc get initial user content?
117 points by moretai on Nov 8, 2017 | hide | past | favorite | 63 comments
If they need user generated content to function, how do they get started? Do they just get their friends to start using the service?

For reddit, all the content was submitted by Steve and Alexis at first, then they got a couple of friends to start submitting.

To make submitting for themselves easier, Steve made a special submit page that would also create an account at the same time, and then make the submission from that account.

After a while they noticed people they didn't know were submitting, and then PG mentioned them in an essay and things started taking off.

They knew they had something the first day they didn't have to submit anything.

About Reddit, they tell the story in this live episode of NPR's How I Built This: http://one.npr.org/?sharedMediaId=545635014:555389162

Why does NPR have so much good content on stuff that I am looking for?

Just the other day on HN on an article about Netflix, someone posted a conversation with someone who works at Netflix from planetmoney. https://www.npr.org/sections/money/2017/09/13/550793717/epis...

I need to delve into npr.

NPR seems to prioritize quality over quantity and clickbaiting.

Because that's their business model.

Wow, every news outlet should become a "publicly funded non-profit membership media organization."

I enjoy hearing Reddit's story from time to time. It's a nice reminder on the hustle it takes when starting a company.

PG rolled up his sleeves personally as well, if you look at his account [0] you will see him contributing very actively in those early days.

[0] https://www.reddit.com/u/Bugbear

Did the founders create sock puppet accounts early on, so new users would get more comments and votes, and feel like the community was bigger and more active?

There was no commenting at the time, but yes, there were sock puppets for voting and submitting links.

I started a forum by creating a bunch of accounts, talking to myself, and pampering my first users with my menagerie accounts to make the community seem like a small charming group even though it was just me.

I was 17, my forum eventually took off, but I don't think I'll ever have the energy for that again.

engineering user traction on social forums or community is challenging and I could attest that it most certainly starts out with engineering "chatter"

now I find it's much harder to do because of reddit and large dominating websites

As someone who's run community sites for a while, there are a couple of ways these sites get the initial user content. Here's a brief list:

* They post all the content themselves under their own name, until other people start joining. It's not elegant and may not look great at first, but it's worked before.

* Fake accounts are used, with the founder(s) pretending to be a number of individuals. Said fake accounts usually get retired as the site gets more popular (for real).

* Previous friends or contacts are involved. Perhaps they run a company and everyone employed is told to post, maybe they're a celebrity with a fanbase, you know the drill.

* Influencer marketing is used. In other words, they try and incentivise celebrities and popular figures to use the site, assuming their fanbase will follow.

* Promotion sites or services are another option, with a few admins doing 'exchanges' for a bit of extra activity. So you go up to someone else running a similar site, and say that if they post on your site, you'll post on theirs for a while.

* They could also just pay people to post. You can hire random freelancers to post content if you're desperate, or use a paid service where people earn money for signing up and posting X amount of content on so many sites.

* If they're really interested in morally dubious tactics, they can also use tools that imports content from third party sites en masse. Seen a few do this with Yahoo Answers, and I wouldn't be surprised if this stuff existed for other services too.

* They could also buy another site (or twenty) and merge them in to get the appearance of a large community. Seen that happen a lot with tech and gaming sites.

* Finally, they could always just directly incentivise people to join by say, offering money to whoever posts X amount of content or refers lots of new users.

Really, it depends on the site in question. Reddit was initially built on fake accounts, but that's only one of about a million ways such sites could take off.

These Quora answers give a good hint: https://www.quora.com/How-did-Reddit-get-initial-traction

In particular, it seems like two big factors were:

- Having influencers (e.g., Paul Graham) onboard, which gets their followers

- Auto-populating content (fake users) and generating the illusion of activity so that it doesn't feel like one is entering an empty forum

Getting friends to use a service also makes a lot of sense for social networks that incubate in small communities (e.g., Facebook), but are a bit harder to pull off for these diffuse communities like Hacker News.

> Auto-populating content (fake users) and generating the illusion of activity

The other owner of a project of ours once suggested doing this. I rejected that because of ethical reasons. I also told that could mean legal trouble. Was I completely naive/wrong?

It's certainly not illegal, at least in terms of social media. There's no contract/obligation/whatnot that a username be mappable to a single distinct human.

I wouldn't really consider it unethical in the reddit case either, but I could imagine opinions vary.

How’d the project go? Did you get any users? I’m going to go ahead and guess the answer is “no.” So yes, you were naive. Very few social apps (or marketplace apps even) overcome the chicken/egg problem without some form of faked content. After all, who’s going to post to an empty forum?

I’ve growth hacked an app to 600k users across 200 communities (colleges). What worked for us was seeding each new community with enough fake content for the first few days. After that the real users took over. The most interesting aspect of that growth hack was that users would post content similar to what we seeded it with. So we could create the “vibe” of the community with fake posts, and soon it turned very real.

You guessed right. At one point I lost my motivation and took the offer to get paid for the work I've since the beginning put in, for my shares. They tried to do the growth hacking stuff but for them it was a thing they read a few blog posts about, so it didn't help. The whole thing just shut down after a few months.

But that doesn't answer the ethical question. Success and imminent failure are not a justifications for unethical actions. If success cannot be achieved without unethical action, then perhaps it is best to fail.

Misrepresenting the nature of the content would certainly fail Kant's deontological tests.

I’d like to hear an example of a chicken/egg social app that did not fake content at the beginning.

If you can’t think of one, or there’s few examples, then there’s also the ethical consideration to your stakeholders (eg investors). Is it ethical to forego a legal opportunity to grow your user base? Do you have an obligation to your investors to do all you can to build a successful product?

My personal ethics revolve around the question of “am I hurting anyone?” I really can’t imagine how posting fake content to your own forum is hurting anyone. By I can see how neglecting to take steps to grow your user base could hurt people (yourself included, if the business fails).

Would you rather that Reddit had never faked content? More likely than not, if they hadn’t done that, Reddit would not exist today. So isn’t that a net positive from an ethical perspective?

The weakness in the teaching of ethics, and in particular the lack of discourse in ethics amongst business in this nation is utterly frustrating. That you can do something (within the law) or that no moral outrage would occur is itself not a measure of ethical action.

Not dealing harm is itself a rather weak ethical stance. Ethics doesn't test whether you do no harm, it tests whether you maximized the good -- and not the good for you or your investors, but rather the good across all of society.

If anyone holds you to an obligation that entails that you take an unethical action then such a license itself become immediately void. We can see this historically expressed by St. Augustine in the 3rd century and more recently by Martin Luther King Jr. (an unjust law is no law).

Hence, the law itself is not a measure of ethical act. That you _can_ do something is not itself a justification of ethical action. Only that you _should_ do something is itself a measure of ethics. In which case you must provide proof that you _should_ misrepresent yourself and your content. Returning to Kant. If you cannot universalize the action -- that is to say, if you cannot show that to create fake content is always a good act (and with all the rufus about "Fake news," I think we certainly cannot show that fake content has been universally good).

Jumping into the question of net good we skip from deontology to utilitarianism. Yet, utilitarianism recognizes that net good is not a measure of money, or pleasure but can be an abstract measurement. Success of the business is only a local consideration, you must expand the ethical question to all of society. That which is good for you or your investors is not itself a net good for society (nor from a deontological perspective a good unto itself).

Was the success of Reddit a net positive for society? Had there been no Reddit would our net quality of life have been improved as a whole society. In this, I would say Reddit was not a net positive for society. It consolidated a distributed system of smaller social forums across the internet into a single hegemonic gated community and consolidating social power into a smaller sum of hands. And I would argue that consolidation of power into smaller groups (whether hard or soft power and in this case we are discussing the soft power of cultural norms) is itself unethical.

It's so clear that the person you're responding to does not even understand what "ethical" means. This is usually the point where I give up on hn threads.

Hey, I have a friend working on a very similar app to the app you worked on, just messaged you on LinkedIn.would love to talk about it

I'm interested in hearing more details. You ever write blog articles that go into more depth?

At some point I will (the story is pretty crazy) but for now I’d like to keep certain details of my life private, so I avoid blogging. I used to think I’d need a blog to attract consulting clients, but I’ve had no problem without one so far, and thus I haven’t gotten around to it. My personal website is basically Lorem Ipsum lol.

If you’re really interested, shoot me an email. I’m happy to talk one on one.

I think there's a difference between providing your own seed content and fake users. In the case of a forum, I see nothing ethically wrong with having a user role marked "Staff Writer" or some-such.

I think there's a bit of a difference between priming the pump so that your forum looks active, vs. having sockpuppets included in your active user numbers that you use to try to get funding. The former's a bit murky, but I'm okay with it at the start of something new and being phased out. The latter's straight up unethical and (IANAL) potentially illegal.

>I also told that could mean legal trouble.

I would've certainly asked for a clear explanation of this as it's hard for me to believe it's remotely true.

In the case of reddit I heard, that - at least in the very early stages, the creators themselves provived a lot of the content, and even hired some people almost exclusively for that purpose. I can't remember where I read it though.

Part of the Udemy course Huffman was offering he said this [1] And honestly, it doesn't bother me all that much. I see it like an engine needing a big spray of gasoline and a spark before it can run. Just because it didn't start with pistons running doesn't mean it's a bad engine. I see it more as faking it till they made it, and it worked. They were their own biggest fans.

[1]: https://youtu.be/zmeDzx4SUME?t=30s


Yeah, they had a post form that had a field for username that was available only to the founders. So they could post content and then invent a username to make it appear that there were lots of unique users. They discuss it in this podcast. http://one.npr.org/?sharedMediaId=545635014:547386946

HN started small, as you can see from the earliest 5 posts:






It essentially started as a small place to share links for the early YC attendees and slowly grew from there.

Amazed to see jacquesm's comment. I knew he was an old member, but it seems he's been active right from the get-go.

Note that his comment was posted quite a while after the submission (although, still a long time ago!)

I also just realised my account is a little older than jacquesm's O_o that.. was unexpected.

They even had trolls and shadow banning back then by the looks of things.

I think when anyone starts anything, they have to begin with their own voice.

I started http://www.confessionsoftheprofessions.com about 5 years ago... I must've written about 10 articles before I even began the website. I was the only one submitting content to the website for the first few weeks, but eventually I ran out of ideas, and started soliciting on Craigslist, and even paid a bunch of people on Fiverr to write articles. I moved to MyBlogGuest and MyBlogU which were free sources of people looking for a home for even more articles, and had been publishing up to 3 articles per day.

After about a year or two, I began receiving at least a half dozen emails a week from individuals, freelancers, marketers, and universities, including lawyers, doctors, construction workers, police officers, teachers, and people from all different professions wanting to submit their confessions (articles). Emails have only increased, but I've got a good handle on everything.

It has become a daily routine in my life and I've spoken to people from all over the world, including India, Philippines, Australia, South Africa, Tunisia, Europe, Canada, etc.

I still write my own articles as I've always got ideas, but I've always got new people discovering the website or repeat contributors who have been submitting articles to the website for years.

Even those times where I feel like quitting and no longer keeping up, I can't let my contributors down, so they keep me motivated to keep it going.

It's nowhere near the size of those but I have run http://www.rubyflow.com/ for 10 years and it was basically a month of me posting several things a day and mentioning it on my then-popular Ruby blog before there was a tipping point where other contributors kept it going. I rarely post at all now.

On sites that require a social aspect rather than just link sharing, I imagine you really need to have a core group of people commit to engaging with each other before others are attracted towards it. I recall early HN being like that at least.

Love Rubyflow. I was just on that site before surfing over to HN.

Reddit's co-founders populated the content themselves. This has been mentioned in interviews with both Huffman and Ohanian and can usually be found in write-ups about "Growth hacking."

They fake it 'till they make it.

By doing jobs that do not scale. Such as submitting the content, and answering the questions submitted.

By doing things that do not scale, you are able to develop a strategy faster since you are in the trenches seeing what is most likely to fail.

PG already had a following because of his essays. He shared a link to Reddit in one of his essays. This allowed them to transition from artificial content to content from real users.

If you're goal is just to copy an existing established site, you're going to either have to generate a lot of content yourself or do a ton of advertising. But, if you're creating something truly new and unique that people will love, then you'll likely have no problem getting a community to come build around it.

E.g. Back in 1997 I built a chess website for playing correspondence chess. There were lots of other sites for playing correspondence chess, but none with actual enforced time controls, and even the weakly/non-enforced time controls were very slow (like 1 move every 7 days). Mine was the first where users could choose their time controls, and the time controls were automatically strictly enforced. The only advertising I ever did was when it initially launched was to post in a couple of USENET groups and forums, and it took off from there. Within a few years sites with deep pockets started popping up with the same idea,and the sheer number of sites that do the same thing today make it impossible for a site with no revenue like mine to get any attention. But even with my almost no advertising, it still got to about 15,000 active users at its peak, and still has several thousand today.

I started Le Journal du hacker [1], the French-speaking Hacker News-like website 3 years ago. I was mostly alone for one year feeding the website until the first regular contributors (now co-founders) showed in.

Good things come to those who wait (and work hard).

[1] : https://www.journalduhacker.net

Reddit, HN, facebook, etc all had the founders generate content via lots of accounts until the sites became popular enough that they didn't need to generate content anymore. Reddit founders even discussed it on many interviews and of course on reddit itself. I remember years ago, they talked about how they created a bunch of content and fake accounts in order to generate interest and to make it seem like reddit was more active than it was. Essentially, in the beginning, most of the content and most of the comments were the founders and their friends submitting and responding to their own comments with a bunch of sock puppet accounts.

Essentially, it's fake it til you make it.

I think I came up with one way to kickstart a site like that here: https://engineered.at/

It's an HN/Reddit-like front end to essentially an RSS aggregator. Granted, traffic is still pretty low and I haven't had time to hack on this a bit, but it does technically solve the chicken and egg problem of needing content for a community that is meant to produce content.

I've always thought you could set something up like this to make it easier to find links to post to more "pure" content sites like Reddit from the onset of the community until traction is found.

That’s not enough. I’ve been posting crypto/security links here https://www.cryptologie.net/links for 2 years and for almost every day and I've been unable to gain any traction. Of course I don't mind because I'm actually following the news and reading a lot of these links.

You are doing good. Now go and post some comments, get other accounts to post comments. Add a summary of the article, maybe use an application that generates an automated summary.

Create some long tail words and through those into the comments and auto summary on the page.

That should rank you a bit for some words in your group. Right now, Google is thinking you are just providing duplicate content.

> Now go and post some comments, get other accounts to post comments

Yeah that's the problem, I don't want to spend my time doing that.

Well tough luck. Sometimes you'll have to do stuff you don't want to do, and when it comes to running a community, being part of said community (or at least faking it until one exists) is one of those things.

I've built many successful forums and chat rooms. You don't get "friends", you usually start off as a fork of an existing community. You first need a group of people who want to talk about something.

I actually designed some of my communities to have sock puppet accounts. I never had to use it, because people just clicked and used it.

So if you're going to build a fitness community app, you should probably be in several fitness WhatsApp groups or Facebook/Reddit groups and so on.

You fake it until you make it, that is how reddit got started.

Slightly off-topic, but not much, something I always find very funny, in a somewhat sad way are the just launched/published sites including (from day one) the FAQ's (asked by whom?) that besides being often so basic as to border idiocy, tend to remain the same ones for weeks, months or years, whilst you cannot find anywhere answers to actually asked (by you) questions.

In this video Steve, co founder of Reddit, mentions some tips, steps they did, like own account creation and submitting links on the fly...


Follow up question. How often do sites get their start by "stealing" content from others?

Crunchyroll got started by stealing stolen content. They uploaded high quality sub group releases to a low quality player without attribution. This is why there is now a sub "group" called HorribleSubs that is literally just rips off of Crunchyroll.

Depends what you mean by stealing. I definitely remember some large services (like say, YouTube) originally becoming popular based on stolen content posted by users and there's a lot of controversy about Facebook seemingly profiting off users reposting other people's videos on their site, but generally it's not the site creators who are stealing the content.

But yeah, audio and video hosting sites tend to get popular or start out on the greyer side of the law before trying to go legit.

some things like a well written article could be stolen but like i dont know a meme or a top 10 list. less so

Growth hacks, fake it.

Fake users, committing real content

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