[JOURNEY] Starting a SaaS and getting my first customers

Heya, welcome to my Journey,

This will be a journal for myself and keep myself motivated to create a SaaS in the engineering area.

Will be updating this journey as I go along.

Background:

I am not a programmer but I work in a close/related discipline where i review and code from time to time, I know how to code and get things done, but it will not be Linus trovalds level of coding! Obviously I will use AI to help me along and code faster.

My Stack is as follows:

Nuxt as Metaframework (VueJS + SSR, this for me is the best framework there is at the moment, easy setup and autoimports make it quite nice for developer experience)
NuxtUI
DrizzleORM + postgresql for the DB
Stripe

Hosting:
My own baremetal server running on Hetzner


I think that my main issue will be marketing and reaching out to customers, will see that hurdle once I have an MVP running.


The idea of the SaaS is quite simple actually, there is a pain point that we encounter in my day to day job that I have an idea to solve. To not spill the beans, it is a diagram editor with some extra functionalities.

Will keep you posted

Today’s progress :

Today was mostly setup and laying the groundwork.

- Set up the full environment
- Used **Dokploy** for deployment
- Bought the domain
- Started the barebones of the SaaS:
- landing page
- authentication
- basic app skeleton / structure

Next up: start building the core MVP features and get something usable as soon as possible.

All the best for the journey.

what does your saas do ? do you need to have full website or just api can work ? i am asking this because if your service can work with just api you can use rapidapi and apify for testing.

seomaster5 said:

All the best for the journey.

Click to expand...

Thanks!

ravikantrathor said:

Following this journey! Building SaaS without being a senior dev is totally doable with AI-assisted coding now.

Few things that helped others in similar position:
- Start with a "concierge MVP" — do the service manually first, then automate. Saves months of building the wrong thing
- For first customers, don't pitch features — pitch the outcome. What problem does it solve and who has that problem right now?
- LinkedIn outreach to your exact ICP works well for early beta users. Offer it free or heavily discounted for honest feedback

What niche/problem are you targeting with the SaaS? Happy to give more specific thoughts based on that.

Click to expand...

Yes, starting at the moment with an MVP and will move along once it is setup. It is an added functionality to Architecture Diagrams that help with understanding

JohnSmithCPA said:

what does your saas do ? do you need to have full website or just api can work ? i am asking this because if your service can work with just api you can use rapidapi and apify for testing.

Click to expand...

Unfortunately it is not an API, it needs the full website. Thanks!

Managed to get the barebones of the site finished up over the weekend.

The site is actually online now (finally lol). Using Dokploy on the Hetzner box was a bit of a learning curve since I'm not a dev by trade, but it's working solid for now. If anyone is looking for a Coolify alternative, Dokploy is worth a look.

What's done:

• Basic landing page is up.
• Auth (login/register) is working.
• Site is live and accessible.

Now I'm moving onto the "meat" of the applogic. This is where it gets tricky because I'm basically building a custom editor inside the browser. AI is helping a ton with the boilerplate, but I'm having to spend a lot of time debugging how the data saves to the DB.

The plan for this week: Focusing 100% on the internal functionalities. I want to get to a point where I can actually use the tool for my own work tasks. If I can't use it, I can't sell it.

The "ins and outs" of the editor are taking longer than I thought, but thats the game I guess.

Still haven't started any outreach yet, want to make sure the app doesn't crash the second someone clicks a button first!

nuxt is a solid pick but that's where a lot of non-traditional devs trip up — the frontend framework is the easy part, the backend logic for billing, auth, and multi-tenancy is where saas projects die quietly. what's your plan for payments and user management, because that's going to eat more of your time than the actual engineering feature you're building.

dirtyFunnel said:

nuxt is a solid pick but that's where a lot of non-traditional devs trip up — the frontend framework is the easy part, the backend logic for billing, auth, and multi-tenancy is where saas projects die quietly. what's your plan for payments and user management, because that's going to eat more of your time than the actual engineering feature you're building.

Click to expand...

honestly the multitenancy is mostly handled by Drizzle + Postgres schemas, so i'm not too worried there, but Stripe is definitely going to be a headache i'm saving for later lol.

right now i'm just trying to make sure the app actually works for a single user before i worry about how to charge them.

Quick update for everyone following along.

Finally hit a big milestone: The MVP is more or less finished. The core editor is working,, and the "extra" engineering features I was worried about actually feel pretty smooth now.

Current status:

• MVP: Done. It’s not "pretty" yet, but it’s functional.
• Bug Squashing: Spending this week just clicking every button until it breaks. Found a few nasty ones with the node rendering on mobile, but getting through them.

Switching gears to Marketing:

Since the app is usable, I'm starting to shift some focus away from the code. I'm currently setting up a topical map for the blog/knowledge base. The goal is to build some "topical authority" in the engineering niche so I don't have to rely 100% on paid ads or cold outreach later.

nuxt is a solid pick but curious if you've thought about the auth layer yet, thats usually where non-traditional programmers hit a wall fast and it can eat weeks if you're not careful with how you wire it up from the start

I am using better auth.

thatadslife said:

nuxt is a solid pick but curious if you've thought about the auth layer yet, thats usually where non-traditional programmers hit a wall fast and it can eat weeks if you're not careful with how you wire it up from the start

Click to expand...

Solid stack honestly, Nuxt + DrizzleORM is underrated for indie SaaS. Building around a real pain point you live daily is the best starting position , marketing gets easier when you already know your customer. Ship the MVP fast and dirty, don't over-engineer before validation.

I've also been researching SaaS projects recently, bought a few frameworks, and am trying out Vibe coding. I hope to have the opportunity to work with you on this.

Some projects are very successful, some fail; I wish you success.

SEOCodeLab said:

honestly the multitenancy is mostly handled by Drizzle + Postgres schemas, so i'm not too worried there, but Stripe is definitely going to be a headache i'm saving for later lol.

right now i'm just trying to make sure the app actually works for a single user before i worry about how to charge them.

Quick update for everyone following along.

Finally hit a big milestone: The MVP is more or less finished. The core editor is working,, and the "extra" engineering features I was worried about actually feel pretty smooth now.

Current status:

• MVP: Done. It’s not "pretty" yet, but it’s functional.
• Bug Squashing: Spending this week just clicking every button until it breaks. Found a few nasty ones with the node rendering on mobile, but getting through them.

Switching gears to Marketing:

Since the app is usable, I'm starting to shift some focus away from the code. I'm currently setting up a topical map for the blog/knowledge base. The goal is to build some "topical authority" in the engineering niche so I don't have to rely 100% on paid ads or cold outreach later.

Click to expand...

congratulation on finishing your mvp. wishing you success with your project

Update: June 23, 2026

Hey everyone, time for a quick update on where things stand.

I know it’s been a minute since my last post, but a lot has been happening behind the scenes. The good news is that everything has finally been smoothed out. All those annoying bugs I mentioned last time especially the node rendering issues on mobile have been completely sorted. The app feels stable, fast, and ready for actual usage.

With the technical side mostly locked down, I’ve officially switched gears.

Current Status:​

• Product: Stable MVP.
• Paying Customers: 0 (Yep, still at zero, but that's about to change).
• Current Phase: 100% focused on Marketing.

The Marketing Strategy: Double Down on Reddit​

While I initially mentioned building out a topical map for SEO (which is still running in the background), SEO takes time to kick in. Since I need immediate validation and real users breaking things, I'm focusing heavily on Reddit right now.

Because this SaaS solves a specific day to day pain point I personally experience in engineering/architecture diagramming, I know exactly which subreddits my target audience hangs out in.