Ask HN: What is the cost to launch a SaaS business MVP
I interviewed several entrepreneurs, I noticed several spent so much money on developing their product before launch. What is your experience?
It costs whatever it does to build something in which you can have someone pay you for and receive value in exchange. Nothing more.
That value is not a SaaS. You don't need build out a "teams" feature or "multiple accounts" feature; you don't need to even build out recurring billing, seat management, or whatever combination of tiered usage you can think of.
You need to give the ability to receive value for something.
I see so many people burn themselves out on doing so and they never talk to customers because all they're doing is fixing edge cases in their billing features.
Once you have your idea's guts, build out the most bare minimum payment model: accept their credit card, store it, charge it, and make a reminder that you need to go into Stripe/Braintree/whatever and charge their card again in a month. Manually create an invoice if your provider can't give you an invoice for them for some reason.
Once managing your customers manually becomes a chore, you can then start to invest in automating it and building out the "boring" infrastructure that has nothing to do with your core idea. Granted, there are SaaS starter kits all over the place for many languages and frameworks. There are also platforms that handle all the recurring billing/seat management stuff for you, too.
But the odds are you might even not get there. The time you would spend building out that infrastructure is not valuable to a customer, and so it's not valuable to you.
Your chances of succeeding are not as closely correlated with subscription management as they are learning about your customer's needs. Take the time you would otherwise invest perfecting a SaaS setup and invest that into talking to customers.
Build your idea, forget that you're building a SaaS. Build the SaaS-y part of the SaaS after you have a customer. Without customer, you're missing the S in SaaS: service. All you have is software.
Build the software.
One way to save significant time/money is by using a SaaS starter kit. There's one for just about every stack these days: https://bullettrain.co (Rails), https://divjoy.com (Node/React), https://saaspegasus.com (Django), https://spark.laravel.com (PHP/Laravel). Ideally you're not wasting any time on this boilerplate.
Otherwise it costs you many hours of market research to see if your product is viable. Many hours of design and testing to determine if your product is useful/liked by customers.
In comparison, the running costs are orders of magnitude less than what you will waste building and launching the product.
However, the product is for ourselves and what it avoids us down stream is huge.
Machine learning projects are expensive and seldom succeed, and this is taxing on morale, retention, hiring, etc. and our platform addresses these problems.
In other words, spending time to make a wheel means it's time you're not spending moving forward, but a round wheel is much better than a square wheel and you make it up in no time when you have it.
Huge costs, but huge benefits that compound.
Early feedback from actual customers on an MVP can save lots of development time. GitLab Service Desk is one way to handle emails as issues from users who don't have GitLab accounts.
A beta invite program / mailing list signup page costs very little to set up; you can start building your funnel while you're developing the product.
Less than $1k if you also incorporate in a standard way, e.g. stripe atlas.
Less than $10k if you also need a lawyer and/or CPA for taxes.
I can't say the cost for non-technical founders, i.e. outsourcing development. I'd guess with $10k you can already get something simple up and running.
Edit: of course I'm assuming cost of your team time = $0.
What you're really asking is "How to create a SaaS app?" which is extremely broad.