Ask HN: How would you launch a privacy-first, Instagram-like social network?

> by far the most common mistake startups make is to solve problems no one has. - Paul Graham [1]

I think this proposal points at an interesting issue that I see crop up often on sites like HN. It goes (I think) like this: 1) "I care deeply about what I perceive to be a problem." 2) Extrapolates onto some large critical mass of people, 3) That critical mass of people does not actually agree with the problem statement in any way, 4) Build a solution, never gets meaningful traction, misdiagnoses the root cause of the failure.

I'm not saying this to dump on this idea. Rather, I think it is a meaningful bias that all humans are vulnerable too. But I do think that filter bubbles amplify whatever this bias is to a powerful degree. If you are surrounded by people who are obsessed with privacy, it seems like everyone is inflamed by the economic model of social networks. But, outside the small filter bubble, most other people don't care at all or actually think its a great model.

Having said all of that, maybe a way to test whether you are experiencing this bias is to interview or collect information from as close to a random sample of social media users as possible. The more random the better. If you find yourself talking to people in SF/Seattle or people in the tech industry, that's a sure sign you have a bad sample. When you talk to them, don't ask leading questions that will bias them. Try to understand what their unbiased views on the problems (if any) are with social networks. Maybe you ask them what their best friend thinks so as to try to sidestep preference falsification. If you discover a consistent problem, maybe you're onto something. But then, you're still up against a world saturated with social networks, and convincing people to pay for something they get for free will be very, very hard. I think, ultimately, the only way you'll discover whether the problems you may have found are really important or not is if you can convince anyone to pay you. I suspect it will be a Sisyphean endeavor.

[1] https://paulgraham.com/startupideas.html

Why would you build it?

That isn't being snarky, it is the question you need to answer. Why would people join it? What problem are you solving for them? Why are celebrities and influencers on those platforms to begin with?

When you can detail out those answers, those answers tell you what you need to do to launch. You have to match the table stakes of existing platforms, which means you need to know not only what those are, but why they are important to the userbase.

I suspect you'll find that your drive for privacy-first is going to be well-received in theory, but not a match for why people join such platforms. But you'll need to do the research to figure out such things.

Privacy first, social network without ads means users pay. Why would any users pay for this? The vast majority of people on social networks do not care about privacy. They care so little that their accounts are frequently hacked because they are using the same password everywhere. These people also aren't going to pay for something they already get for free.

Your motivation is purely revenue. What would the user's motivation be to add yet another thing they have to keep up to date on? What can any users you do get use to convince their circle of humans to also join?

Without massive user counts or revenue share with celebs or influencers what incentive would they have to join?

How are you going to pay for bandwidth and storage of videos/images with no revenue? Like do you have a nest egg/run rate saved to fund this social network for a year or two?

Apple photos is my privacy first social network. I have "shared albums" with my families, different friend/hobby groups. There is an activity feed so whenever someone shares a photo I get a notification, can like like, comment. It just has the basic features and works really well. Here's the thing, I pay $40 a month for the icloud bundle. So technically they've monetized it already. I've always wondered if others use it similarly...

There exists a federated Instagram clone in case you want to look at previous attempts:

https://en.wikipedia.org/wiki/Pixelfed

It's called a private group chat, and it exists everywhere. No "recommendations," no ads, no moderation, no randos butting in trying to derail conversations, just shared social norms among the participants. It's not something you can scale or monetize, and that's exactly why it's the perfect social network.

I wouldn't bother. Nobody really cares about privacy-first. The HN audience is one that you might expect to care strongly about privacy, and yet even here many still prefer to use the Android/Google ecosystem because it's more customizable and Not Apple, over the highly private Apple ecosystem. Privacy just isn't that many people's first concern.

Pixelfed is already what you are looking for. You could contribute yourself, fork it, or just use it in your server.

What I don't see often, and I think it was a neat idea is a copy of Path. A social network for a very very small set of people to share their things. I think path was like 2 members.

I imagine that just a whatsapp/telegram conversation chat is more than enough for this case tho.

As a new service, I think you'll struggle to attract celebrities and influencers. You'd need to have some really compelling reason for them to use the service, and the only compelling reason they'd care about is audience size, which you won't have.

If I were attempting this, I'd design and aim the network for a niche audience of some sort and once that's established, begin expanding to wider audiences.

Most people don't care about privacy, and those who care – like a lot of users of this website – care too much to trust you.

I don't think starting with celebrities and hoping you instantly become Instagram is the move. Do something different, and better, than Instagram that some group of people will want... privacy isn't it. You prioritize privacy, I'd even say you should, but that should be an added benefit, not your primary sales pitch. As much as I like it, the public at large has collectively responded that they don't care enough. So what novel thing will you have that will draw in the kids and make it cool? And remember that no matter how good of thing you make, there is a large element of luck in becoming an Instagram. The second you describe your product by saying it's like another existing and popular product, you're already at a disadvantage. There is no shortage of instagram clones.

I wouldn't.

I tried and tried to sell stuff to "privacy freaks". Turns out that type doesn't like paying for closed-source services and generally will turn their nose up to subscriptions. I am one of these "privacy freaks", love them, but they are honestly probably the hardest people to sell to.

I stopped trying to market to them. It is also a nightmare to try to reach that audience too because they almost always have ad blockers and any space they hang out in very quickly sniff out ads or outright ban talking about products they created.

Furthermore, privacy people tend to be technical themselves so they would prefer to build a solution specific to them rather than use someone else's solution unless it is difficult or requires specialized knowledge they don't care to acquire.

I'd be more worried about how you're going to pay for it than anything else. If the revenue is not from ads how are you funding it? Especially if it becomes a hit.

I think you'll have a hard time convincing most users to pay for something they can get for "free" elsewhere.

Path??? https://en.m.wikipedia.org/wiki/Path_(social_network)

Many artists (and other users) would like to leave Instagram because they don't like Meta using their work and posts for AI training, but doing so is difficult because of how popular Instagram is compared to alternatives.

https://wired.me/technology/artists-are-leaving-instagram-ai...

> One challenge I'm anticipating is attracting celebrities and influencers who are already established on Instagram and TikTok.

I would rather like to see the solving the problem of not attracting any celebrities or influencers , or companies and just getting normal people for social connections, like Facebook was originally. This is the hard problem.

Money and fame usually destroys the platform in the end, as it aims for endless competition and rule breaking.

As others have already stated, without either actual paying users (not the celebrities) or some form of advertising costs can become very high quite fast.

Even though you are against all forms of advertising, some really privacy oriented form of it may be acceptable by your intended users. Some thoughts:

- Users need to disclose some topics/subtopics (either broad or even very niche interests)

- The ads need to be conformant to some predefined, non-intrusive style guide: maybe black-and-white or a limited color palette and font selection. This could lead to interesting and even creative ads on the platform

- Advertisers could get anonymized click-counts for their specific categories, and maybe some other category-based info, but user data would be immediately aggregated and not used for specific targeting

- Base users would see 1/3 ads in their feeds, but some payments could reduce this to 1/5, 1/8, all the way to zero.

This approach might provide the intended privacy benefits with much less noise, but also not alienate the average user that expects social to be free of any cost.

I think you are worried about the wrong problem. Forget about the influencers for a moment - what will your source of revenue be if you don't have an ad network?

Users are the content on social networks. If you charge them to join, you will have fewer users and less content on the network. That reduces the value so you can't charge as much, and the cycle continues.

Many commenters are telling you not to do this, but no one is answering the main question - how to launch such a network. I've actually been thinking about this idea a ton... how could you ever get a niche social media network off the ground when the incumbents are massive?

In any case, I'm working on a small project to support smaller social medias. Not trying to promote too much but it's called Soshials (soshials.com) and is basically like producthunt but for social media platforms. An additional twist is that users can actually join "cohorts" to try the new social media platforms for a month and give feedback.

I think getting feedback from real, initial users (instead of just friends/family) is a real reason why it's hard to iterate on and improve new platforms. Interested to hear if others have similar thoughts on how to actually launch a new social network.

Hopefully you make an amazing product and I’m rooting for you, but privacy is probably not a killer feature for social media.

I’d focus on finding an amazing new way to rethink how usage works and privacy can be a nice afterthought. But know this is possibly the most competitive consumer software space in existence.

Imagine you have asked a question of how to get from Vancouver, B.C. To Hudson Bay. In response you get answers like "Why?" and "Here are other people who have failed", "People have already been to Hudson Bay", "Money and Power will destroy you", "Don't take camels".

And there is your problem statement. There is vast uncertainty about how to get there, even what ways there are to get there. You have to clarify all the uncertainty and there are no a priori answers. Just start going and see what happens. There is no recipe and if there was it would probably be for something you would not like.

[Edit: I am serious, not trying to be critical or facetious. As someone who has tried similar things, there is no recipe, so if you want to do x you just start with that fact and go.]

Social networks are social advertising networks.

A privacy-first solution by its nature is unlikely to benefit from network effects. You don’t organically discover content that has to be explicitly shared with you.

So they way to do it is to already have a network or platform and then allow private photo sharing. E.g. Apple Photos built on the iOS ecosystem or if WhatsApp had a photo feed.

One idea would be to sell photo-sharing tech to a non-US messaging app that’s popular abroad like Viber or CacaoTalk or that serves some niche community. (Photo-sharing for low-bandwidth users in developing countries?) Maybe it aggregates shared photos into an ergonomic feed or something. But they could just copy that themselves; it’s hard to monetize an improved UX directly without the data/community to power it.

The concept of "privacy-first" and that of a social network like Instagram or TikTok with celebrities and influencers seem incompatible to me.

The entire point of such a social network is to be public, exposure is the point, not privacy. You want to broadcast your content to the entire world, you want recognition in terms of "likes" or for the luckiest, money, you don't go there to keep secrets.

For a privacy-first social network, usually what people want is to form smaller groups where people know each other. Think more like Discord than Instagram, or for a more privacy-focused alternative Matrix.

Alternatively, maybe take a look at OnlyFans. They are quite successful and don't depend on ad revenue, and considering what they do, I think people care about privacy.

I wouldn’t.

If you were really going to approach it, you would launch a privacy focused messaging app, which could add filters and other simple Photo and video editing tools because most people are now sharing in small message groups and not publicly for everyone to see.

I see many takes in the comments but no one seems to be addressing the fact that privacy and social are opposites of each other. I think you could state the problem better- for example sharing with the audience that you expect to share with or some better controls on who sees what. But these alone are not good enough differentiation and in theory Facebook has very fine grained privacy controls and no one bothers to set this correctly and advertisers still see more than what you set etc

It is a hairy problem but it’s hard to frame it in such a way as to make it attractive enough to the masses to switch from existing networks

I'd build it on nostr. You have no infra costs, there already exists a network, and you can build whatever kind of client (privacy oriented in your case) you want. You can also extend the schema as you propose/see fit.

Thank you for your post. I've been working on a similar concept but for super close-knit groups, think families, college friends, etc. (less than 10 people/group). I was going to incorporate some mechanisms to encourage posting since if no one posts, it'll die.

I felt like this is a problem I have. I want to hear about my friends’ lives, but no one posts on Facebook/Instagram anymore. And my feed is now constantly influencers.

However, after hearing all these comments/experiences, I will be pivoting! Thanks you for your post and saving me a headache.

Instagram is for self-promotion and/or what we could politely call "face posting". It's all about showing off and associating with real people. It's basically a personal brand platform.

What does privacy-first mean? Don't socials already have two modes: public-post and friend-post? Celebrities are public figures, why would they want a privacy in this scenario?

I would start simple. Build it and share it with your friends and family. If you are in a college, share it with your college. Solve the problem for your closed group first. If you can convince your group to pay, you have crossed the first step. So first degree of separation is solved.

Remember, most celebrities will be at most at a distance of six degrees of separation. So you can build up to that.

Try taking foreign investment from authoritarian governments and using that money to form exclusivity contracts with influencers on your platform.

I worked on something very similar: a privacy-first Instagram clone built on top of Matrix. We had a prototype and some initial users.

Here's the unfortunate truth:

The challenge of overcoming the network effect is _the_ problem you're solving here. Nothing else really matters. Almost nobody will care that it's privacy-first. Asking HN to help you figure this out indicates to me that you should do something else.

I suspect that the primary challenge social networks face going forward is adversarial AIs. This can take the form of spam bots: astroturfing, trolling, abuse. And of AI spies: bots that programmatically analyze and extract information from users.

I don't know what the next big social network will look like but I think the one key thing you need to solve for is authenticity.

>privacy first

>social media

you don't. you just killed your only decent revenue source, and will never out compete others. Social media live and die by their users.

What you are describing sounds an awful lot like Snapchat. I would honestly just download Snap and tour their App since they have already come up with solutions to all of these problems. They aren’t always great solutions, and snap is still mostly an ads platform for revenue, but the core product gets right to this issue.

I’ve built a social network that has an influencer -> audience model in order to not rely on advertiser revenue. Im trying to figure out the getting users but now.

I’d suggest that the answer is not how to attract celebrities but in the first instance how to attract any users. My platform is novel and interesting and even then it’s not easy.

I've had a similar idea - my thoughts would be to make a distributed p2p social network where data is held on the user's device, and the user could opt-in to selling that data to various sources.

That said, I have no idea how to even vet the feasibility of a network like that, but one can dream.

It already exists, and it's called RSS.

I would use discourse to build it. The feature set is incredible.

I would make it community / area of interest centric (like Reddit).

Attract deep subject matter experts for the categories.

I actually own some domains that could be perfect for this site….

Email me (charles@turnsys.com) and I can go into more detail.

If you’re passionate about this idea, launch it! Your initial costs will be on the low end. If it does grow to the point you need to worry about costs, that’s a good problem to have which you can then handle appropriately at that time.

A privacy first social media site is like a health first pastry shop.

> How could I persuade them to use this platform as well?

Without funds to pay them directly, you'll most likely need to offer them equity or some sort of commission/affiliate model.

why not offer two "plans", an advertising plan for the hordes who don't care, which you milk for revenue to support the infrastructure while you grow the privacy subscriber base? work out the tech to chinese-wall the fortune cookies on each side (cookies like browser cookies, fortune for all the money you're gonna make, and because I said chinese-wall even though they are historically japanese)

I don’t know why people keep trying to put “privacy first” and “social network” in the same sentence, it’s clear by now they are antithetical.

1. acquire lots of money 2. pay celebrities and influencers to create exclusive stuff for your platform 3. ...? 4. profit!

What makes it "privacy-first".

It sounds like revenue-first.

Seriously, take that energy and enthusiasm and study business. You're asking the wrong questions.

It’s an oxy, moron.

An idea like this could truly make you a multi-millionaire!*

*one minor catch is that you’ll have to start as a billionaire

You don’t.

If I had a time machine, I’d make something where text posts are free, but need to pay for photo/video storage. Add some procedural graphics for efficient flare. The circles functionality would be helpful also.

Launch? Need a marketing budget and someone who knows how to spend it wisely. Maybe a contest as well.

> I'm planning to launch a privacy-first social network in 2025, inspired by platforms like Instagram

Brain dumping my thoughts in this space (actively working in it), sorry if it is disorganized.

I would start by deconstructing your problem statement more and focusing on _why_ you want to build this; be very explicit and precise about your thesis. After getting a clear and conscise articulation of your why, your thesis, go talk to people about it. Randos at the bar, the gym, your church, school. Your social network, if successful, will be used by these people. Find places to kick up conversations with strangers. Ask them what they do. Eventually they'll ask you what you do in return. Use that as an opportunity to pitch your "why" and see if they get excited.

If random people aren't getting excited, your why isn't compelling.

If you are wanting to imitate Instagram, you are building a _social media_ application not a _social networking_ application. Social networks are focused on community building (creating social connections). Social media is a variable rate reward machine (casino) that hypes up randos on the internet in an attempt to make them famous in front of other randos on the internet. Social media algorithms optimize for things like "engagement" and "view time" - is this compatible with your current thesis? If not, what changes do you need to make to the Instagram experience to make it compatible? If it is privacy first, how will you feed your algorithm?

You also need to solve the game theory of the platform. Its more than just revenue, though getting people to pay you for this is a big part. The game theory includes moderation, lawfare, personal morals, etc. How are you approaching _sharing and storing content_? Are you going to slot a server into a server rack? You'll have to pay for that of course. But you'll also have nation states and other power seeking actors knocking on your door about your thesis and asking for control/influence over that server- what are you going to tell them? How are you going to align incentives so that your thesis doesn't fail under the pressure from U.S. clandestine operations? Will you be compelled by law at some point to abandon your thesis? Will you be compelled by your own morals to abandon it? How will you handle that?

When it comes to infrastructure costs, the more you personally host the more revenue you'll need to keep the thing afloat. It doesn't matter if you are a non-profit, a public benefit non-profit, a for-profit entity... someone will have to give you money to pay for storing and serving all that content.

The more you can rely on the infrastructure provided by your users (their internet, their compute, their storage) - the cheaper it will be to run. The more you rely on your users infrastructure, the fewer centralized points of control their will be for power seeking actors to try to capture.

This is getting rambley, sorry. I've spoken on this before (in the context of building peer-to-peer social networking experiences) if interested: https://www.youtube.com/watch?v=WeZgCy3ub6M

You make a photo app where backing up your pictures online costs a small subscription fee per month. It’s only free if you only use the native client storing your media locally.

It cannot work if it’s a free service unless the entity running it is a non-profit ie a private VC funded entity will eventually sell or share your data.

To my knowledge, only Apple has made something like this but it only works best on their platform which limits it.

Having it be decentralized is more important than anything IMO.

True privacy comes from having control over your data.

Best of luck with you on this endeavor.

Regarding getting celebrities and influencers, you're going to have to pick from a group of people who are either highly tolerant of seeing their content next to, say, something really odious, or are already considered persona non grata. I'm certain you understand the reasons for this, so let me jump straight to my suggestions:

1. Work on a federated model a la bluesky that is censorship-proof but highly customizable so that people can form up around content they want to see and content they one (this is a very hard balance)

2. Be 'free speech friendly' and accept the fact that you may have to censor, but just be transparent about it. One of the biggest problems with censorship on the bigs is that its opaque, inconsistent, and nonsensical and hurts people who have no idea what they're doing or why it's 'inconsistent with Meta's values' or what the fuck ever.

You are opening a can of worms but its also a much-needed alternative so I wish you the best of luck.

Find a niche community that wants something like that, who typically have disposable income. (Photographers)

Make a privacy-first Instagram-like social network and (here's the important part) charge for it.

https://glass.photo/ is a good example.

No e-mail/phone number requirement. New uploaders must be vetted by moderators/quarantined. No video support. Use AI to detect if something is a meme then shadowban the post. Design must look like https://www.fotki.com/

Hi, I built one of these a few years ago. Poured my life savings into it.

You’re about to get a very tough education on who uses social networks and why they do it. It has nothing to do with privacy. Would urge you to not bother.

[dead]

[flagged]