/sug/ - (tech) startups general - festive edition

[Anonymous No.57092887]

Merry Christmas anons!

/sug/ exists so anons can talk about their startups or ideas for startups with other anons.

> What are you working on right now?
> What stage are you at (ideation, validation, seed, post-revenue, ...)?
> What's your roadmap looking like?

If you're starting a tech business, this is the place for you.

> Good resources:

How to interview potential users: https://manuelohan.com/wp-content/uploads/2017/05/The-Mom-Test-en.pdf

How to write a quick business plan: https://www.leanfoundry.com/tools/lean-canvas

pmarca's top posts related to building a startup: https://pmarchive.com/

Various resources for startup founders (books, articles, etc): https://github.com/KrishMunot/awesome-startup

[Anonymous No.57092889]

Sugma

[Anonymous No.57093094]

>>57092887
This is a gambling subreddit, this post is off-topic.

[Anonymous No.57093104]

has anyone ever thought to build a solar farm?

[Anonymous No.57093252]

>>57093104
I've helped build solar cars while I was at university and it was such an intense operation (capex required, how fragile these things were and their general inefficiency) that I decided against exploring that avenue until I get a lot more money. If you're interested though, a lot of countries are starting to subsidise people installing panels in their home and paying them for excess electricity; could be a good starting point to see what that world is like
>>57093094
this board makes me wish crypto never hit mainstream

[Anonymous No.57094286]

bump for santa

[Anonymous No.57094420]

My startup idea is a robot waifu.
Do i tie it to a shitcoin or nft or sonething?

[Anonymous No.57094716]

Cool topic anon. I committed the rookie mistake of building an app nobody actually wanted. Good experience though. I now never do anything in biz without first investigating the real market for it. I'm reading the book you recommended now, "The Mom Test." I have a ton of ideas on my idea board, and this might help me to narrow them down.

[Anonymous No.57094776]

>>57094716
You can still put some ads on it

[Anonymous No.57094818]

>>57094776
My first thought is to research the google keywords numbers, but I don't actually know how to do that. I am currently busy running an engineering and custom fab business, so I don't have the time to just dive right in. Once I validate some of these with the keywords data, I'll make landing pages and shoot out ads. Any thoughts from you, anon?

[Anonymous No.57094855]

>>57094818
I mean its a phone app you can put ads on it?

[Anonymous No.57094872]

>>57094855
I suppose I could. The issue is that it's in an extremely saturated market. It's a ballistics calculator. In terms of accuracy, it's actually the best one out there (I'm an ME and qualified sniper, which is a rare combo and enabled me to understand exactly what was needed in the algo). The real problem with this app is that 80% of gun owners' guns just sit in the safe. 19% just want to make noise. It's only 1% or so that actually care about real accuracy.

I have like 50 users a week or so; can't imagine that would be enough to make real money on the ads.

[Anonymous No.57094882]

>>57094872
Oh you could make a youtube video with you going to the range to advertise it i guess.

[Anonymous No.57094937]

>>57094882
I've basically just given up on it because the market is too small. I'm making real money in my eng and fab biz, but the desire to make novel technology is still there. There are just so many absurd problems still plaguing the world that I can solve.

[Anonymous No.57095036]

>>57094818
In terms of market research and diving into stats before reaching out to people, I'd look into the following:
- Market reports from research firms (find a market that is similar to yours)
- Annual reports from companies in a similar space
- Google Trends & SEMRush for keywords (I personally don't know too much about it but SEO is a great channel for passive growth)
- VC firm reports on trends they're seeing (likely always going to follow bandwagons)
- Compile your own bottom-up research (i.e., what kind of customers are you targeting? How many of those are there? How much do they pay for similar solutions? Ergo, what are your TAM and SAM?)

I don't put too much weight on market research, especially if you feel like you're doing something novel (vs improving an existing product)

Then, I'd start with looking for communities of people that may be interested in your product and speaking to them in a one-on-one fashion (as written about in the Mom Test). I've found Reddit (inb4 >reddit) and LinkedIn to be great avenues for finding these conversations, depending on what you're making.

[Anonymous No.57095055]

>>57095036
Theres google trends but like you said theres a lot of problems that can be solved that are plain to see.
For example how do you sell code without people just giving it away afterwards?
Is there even a way to solve it?

[Anonymous No.57095667]

>>57095055
Typically you'd apply a small layer around your code to hide the 'raw statements' but still allows people to access the logic behind the code
Think of how OpenAI sells their model (which you could represent as code):
1. ChatGPT i.e., an application in front of that code so that people can access the underlying logic of the code without having access to the code directly. This is more suited for a non-tech audience
2. Paid API; you can access the underlying code (i.e., GPT3.5, GPT4, etc) and implement any application you want around it. This is more suited for a developer audience.
There are quite a few ways to build either solution quite quickly nowadays and allows someone to 'sell their code' without exposing the code directly

If you want a more literal way of selling code, you could host most of it on GitHub and then sell a configuration file that's made unique for each purchaser, but I've not seen that model before.

[Anonymous No.57095809]

>>57095667
I've seen some (Odoo comes to mind) that offer the product free open-source, but it's a bit rough. They offer a SaaS paid version. In my company's case, I started with the free version when we were tiny, and when we got bigger I went to the paid Odoo as I no longer had the bandwidth to play with my little Alibaba linux homebrew server.

[Anonymous No.57095865]

This is a good book on how to talk to users.

Me personally, I'm the tech guy building a SaaS product for a partner's idea.

I also have my own idea I've been thinking about for a long time and am still researching to find out if the whole thing even makes sense - it's very technical related to identity and authentication.

[Anonymous No.57095883]

>>57093094
take it from me startups are worse than gambling

[Anonymous No.57096030]

>>57092887
I am trying to make an arduino uno r4 respond to a joystick. I will keep you all updated because for some reason talking through this shit makes it feel less boring.

I had to solder the fucking joystick, so hopefully I didn't fuck it up.

[Anonymous No.57096055]

>>57095809
Oh I get what you're saying now. Yeah there is also a classic play in open-sourcing your software and then charging for implementation, support services or enterprise-level features. Wikipedia has some decent examples for other models too: https://en.wikipedia.org/wiki/Business_models_for_open-source_software

My only gripe with most of these models is that they are service-based businesses which are not as scalable as software but that might not be a dealbreaker for you

[Anonymous No.57096089]

>>57095809
>>57096055
P.S. my website got copied within 3 months of launch. That's to say if you go SaaS, you better be on top of it and stay ahead of the competition non-stop. I think that's why people get funding. Otherwise you get copied by someone who is funded.

[Anonymous No.57096102]

>>57096089
>>57096055
I've thought about doing the open source thing for a project.

Like tiered pricing:
1. Free open source (clone repo or download docker container)
2. Hosted SaaS in my cloud tenancy, paid per month
3. Enterprise plan where software is hosted in the customers cloud tenancy or on prem network

What you're saying makes me reconsider the free open source option.

[Anonymous No.57096185]

>>57096102
That was my immediate first thought when I read your post - I wouldn't recommend it. The free tier can just have very limited resources available, but still from your AWS/Azure/whatever.

> 3. Enterprise plan where software is hosted in the customers cloud tenancy or on prem network
This is fine, because if an enterprise is paying that much, they aren't going to steal your source and the devs there are getting paid enough to be too lazy to do it.

This also makes me put more and more on the backend/API side, since websites are trivial to copy now.

[Anonymous No.57096244]

>>57096102
It's not the worst setup anon. You'll just have to be good at getting and closing enterprise plans, because a lot of your resources are going to be tied up maintaining a non-revenue-generating product.
It's certainly viable but I'd make sure you have some decent capital and a good plan in getting enterprise contracts. Get the open source tech out early to get some traction at the bottom and then sell to people at the top.
In a way, it's kind of similar to free-tier products like Slack

[Anonymous No.57096261]

>>57096244
I gather that you are a bullshitter. Still glad that you made the thread though.

[Anonymous No.57096262]

So you all are just going to watch them keep amputating pieces of my nose every day then?

[Anonymous No.57096271]

>>57096089
I'm sorry to hear that anon, did you manage to stand out from the crowd? I'm also starting to realise the importance of capital and having a team that can accelerate as soon as you have traction; once word spreads of your success there will be 100s trying to imitate you.

That's why a lot of people talk about a 'moat' - what have you built that people can't easily copy?
7 Powers by Hamilton Helmer is a great framework to think about when considering moats

[Anonymous No.57096280]

>>57096261
Why's that?

[Anonymous No.57096308]

>>57096280
>>57096271
A lot of abstract / high-level quotes ideas, when in reality it's just "build something". For example you can tell >>57096102 is speaking from experience. It's broad, but he is trying to solve a real problem, which is awesome to see. Anyway, definitely don't want to start an argument, but did want to call you out.

[Anonymous No.57096341]

>>57096308
>>57096280
And with all of that said. I will skim the book, thank you. You can read Zero to One by Thiel if you haven't already. But again, all of this high level stuff is way too abstract unless there is a good idea that is executable.

[Anonymous No.57096360]

Cool thread. I'm working on making my first game solo and part time. Does that count as a tech startup?

[Anonymous No.57096409]

>>57096360
1. Luigiii, pass the Lugerrr.
2. To your second question - how do people plan to monetize basically doom clones? It's really cool, but I can't see it selling since your competitive advantage has to be story/cool gameplay mechanics. Like Blood+ still sells.
3. How are you doing it? That's some ancient engine, not Unreal, right? Adding your own assets to an ancient engine?

[Anonymous No.57097120]

>>57096409
Its Unity. Pretty easy to get it looking like an old engine, but with all the modern day engine useful stuff.

[Anonymous No.57097176]

>>57097120
Why Unity over unreal with all the bullshit they went into over pricing recently?

---

My update: Joystick inputs are fucking FAST on Arduino. Fuck the Meadow F7 - I wanted to use C# over C, but fuck it, I am converted to Arduino. Fucking Meadow took 150ms to scan its input ports. Working on logging cycle time now so I know exactly how fast it's going before I start hooking up outputs to servos.

[Anonymous No.57097303]

Are any of you considering porn games?

[Anonymous No.57097343]

>>57097303
Absolutely. If AI was there for image gen, I would make a normie "adult visual novel" porn with a good story using RenPy or whatever it's called.

It seems like the money is there. Even without AI, 3d renders seem fine. I don't actually know how much time is put into the models though. Please tell us more if anyone knows.

[Anonymous No.57097808]

>>57092887
My parents aren't rich and I'm not Jewish be real guys can I even make a startup or will any good idea of mine just get stolen. I'm 30 btw

[Anonymous No.57098094]

>>57097176
Unitys new pricing model only affects games made with unity 2024 or later (not out yet) and games making more than a million dollars a year (not me). Unity itself is a much faster engine to develop with than unreal (no compiliation times, doesnt constantly crash, has actual documentation and just works when expected), while still providing everything you need unlike godot (still very in development e.g. no terrain tools).

[Anonymous No.57098313]

>>57092887
my startup idea is kind of like mcdonalds but for americans or something

[floodbringer No.57099423]

Releasing my custom steam project soon, fingers crossed

[floodbringer No.57099435]

>>57099423
4chan has been kind to me so far. it is NOT a game

[Anonymous No.57099494]

>>57099423
>>57099435
Good luck anon!

[Anonymous No.57099659]

>>57092887
sit-n-spin but for adults
app that turns your phone into a mouse using the flashlight + camera
pizza with pineapples, but the pineapples have been marinated in mtn dew
jameson with a splash of peach tea
no one steal my ideas please

[Anonymous No.57099673]

>>57099659
oh and chicken liver food truck

[Anonymous No.57099675]

I have been working on an AI tool suite for a little over 7 months now. Had a lot of traction early on, but the hype died down. One of my products is an extension/website that allows you to summarize any webpage/document with AI. There's also a product for creating a custom knowledgebase and asking questions against it. Problem is there are SO DAMN MANY ai startups raising money right now, and VCs look the other way if you're a solo founder. They say you need a "business guy", which I would happily do if I didn't have so much development work on my plate. Is anyone in the same boat?

This is one of my products btw.
https://summarize.wtf

[Anonymous No.57099697]

building crypto donation app w nft impact certificates so people can see how their donation get used
it has turned my hair completely white, i'm a nervous wreck and i'm afraid the stress is giving me cancer

[floodbringer No.57099754]

>>57099675
sounds like 2 layers of chatgpt shit.