Ask HN: How to succeed with an AI SaaS? (I have no audience and limited budget)

Hello HN,

I’m a 24 year old MBA graduate based out of Canada

The AI space is exciting for me. Mainly Gen AI. I’m passionate about business in general too.

But the truth is my tech experience is limited and I have no following/audience.

Tech: I don’t think of this as a limitation because if my project takes off, I can always hire people or find co-founders. And to begin with, I don’t need a fully developed product. A MVP from Bubble will suffice.

No audience: This is what makes me feel let down. I don’t have any sort of audience to launch to. To make it worse, my budget for marketing is roughly $500 USD. It’s not enough to get me sponsorship placement or test ads. Which is why I feel lost when it comes to marketing the project.

I need advice.

I’m willing to put in the time and effort it takes to make this AI SaaS a success. I’ll be working on this full-time starting April.

I’m targeting founders with funding/revenue and business owners (with a professional management in place). After working full-time for a month, I’ll decide whether to continue or not.

What are my best options to gain traction within a month in this competitive space?

18 points | by DinakarS 30 days ago

18 comments

  • RileyJames 30 days ago
    What is “AI Saas” ?

    If you don’t have tech skills to build it, you’ve got to be able to craft a pitch and communicate it.

    I have no idea what “AI Saas” means, but being that broad there would have to be customers of yours here. At least pitch it to the point where we can declare yes or no.

    And then go pitch and win your first 50 customers. One by one.

    Then go get funding on the back of your 50 letters of intent.

    The whole time you should be looking for a technical co-founder, and you should give them 50% of the business (equals). Without them you will have to give away more equity to raise (because you won’t have any product, or revenue), and you will burn more investment building and iterating your product.

    Anything less than equal will not work.

    You could make convincing arguments about why the work you’ve already done is valuable, and that they’re coming in later so they should only get xx share. But the reality is everything before that point is mostly irrelevant, what will make difference comes next. And for years into the future, if it’s going to be a success.

  • joshstrange 30 days ago
    Um, do you even have an idea yet or just "AI SaaS"?

    Because if you don't I have no clue what you are doing. No technical experience, no technical cofounder, sure that you can throw together a no-code PoC to get investors (ok, this part might work for all I know, investors throw money at stupid things all the time), and you want to do it on $500.

    Either you have, what you think is, an amazing idea or an incredibly inflated idea of what super powers a MBA gives you, maybe a little of both.

    I'm sorry to be harsh but the levels of arrogance/entitlement in this post are off the chart. Also I'll tell you this: an idea is worthless on its own. You have to be able to follow through and that's measured in years not weekends and I find it hard to believe you could do it without any real tech. Then again 90% of the AI stuff posted here is just a thin wrapper around OpenAI's GPT-4 and at least some of those are getting funding so maybe I'm the wrong one.

    • rramadass 30 days ago
      Your response is on point and i hope the OP takes it seriously.

      The post is so vague as to be totally meaningless. In fact this is going to be my poster child for the cluelessness pervading amongst the "AI/ML entrepreneur" folks.

  • asdfzalsd 30 days ago
    Like others have mentioned you need to find a problem first.

    1) Find what people are already paying for (on Fiverr/Upwork) 2) Build a product that solves that problem

    Use https://insanelygood.tools/problems for free market research

  • Art9681 30 days ago
    Identify a problem people need solved first. Then go solve it. You may or may not need AI to solve it. Then make sure it scales and is secure. After all of that, then begin to think about "AI SaaS".

    Understand that the great majority of products you see are not making any money, funded or not, and the reason they are being funded is because rich people play the lottery too.

  • ratg13 30 days ago
    You can’t “target founders” with vague ideas.

    You need to build something worth showing to people before anyone will invest in you.

    Your only option is picking a book to read on how to start a business.

    Your school seems to have failed you.

  • itisit 30 days ago
    What are you really asking for here? You're providing next to no detail to garner any meaningful responses.
  • solumunus 30 days ago
    You’re looking for people to help you develop this idea and you’ve asked on a forum full of people with such potential, yet you haven’t even explained the idea in the slightest. You will arouse nobody with this post.

    If you’re so sure you only need to do a full month of work to come up with a demo that will convince investors, why aren’t you starting with that? I’m really not sure what you’re expecting to achieve with this post.

  • rozenmd 30 days ago
    Forget what you built the solution with. What problem does this solve?

    For early validation you can email/LinkedIn message 100 people you know, and ask if they've been having the same problem, and what they've tried to do to resolve it.

  • p1esk 30 days ago
    I thought the answer to this question is what they teach at an MBA program.
  • AlchemistCamp 30 days ago
    If you don’t have money, or distribution, or tech skills, what exactly do you bring to the table as far as building “an AI SaaS” is concerned?

    Do you have a competitive advantage on any dimension?

  • matthewsinclair 30 days ago
    Is this satire? I genuinely can’t tell if this a serious/sincere question or an attempt to parody the kind of hype-driven (and possibly even AI-generated) nonsense around gen-AI.
    • DecentRecruiter 30 days ago
      Fingers crossed it's satire, or my mind will just shatter.
  • pesfandiar 30 days ago
    Are you going for a quick value capture, or do you want to build something? If it's the latter, you need a cofounder who can actually build it.
  • ipaddr 29 days ago
    Best option to gain traction with a thrown together Bubble app?

    Take the $500 and buy an existing social media site.

  • Spooky23 30 days ago
    Figure out something complicated that people do in Excel. Make a chatbot do it.
  • gtirloni 29 days ago
    Sounds in line with most "AI SaaS" lately.
  • heyjamesknight 29 days ago
    Hey mate, everyone decided to dunk on you instead of actually being helpful. HN is full of nerds that look down on people like you. Don't take it to heart.

    AI isn't magic: it just does things people do, really fast, and at enormous scale. Any "AI SaaS" is really just a productized version of something you can do with your hands—just done 1M times faster by AI.

    Find a thing that's being down slowly today. Learn the ins and outs of that workflow. And then plug AI in.

    If you're open to it, I'd love to chat. Email me at james@nonerds.com

    Best of luck!