DEV Community

Aashir Khan
Aashir Khan

Posted on

How I would build an MVP today

In the past 2 years, I've learned a lot transitioning from just being a front-end developer to being a product manager. My thinking now is very different.

Before I focused too much on technology, and just procrastinated on the things that don't really matter in terms of business and the word "MVP". 

This is how I would build an MVP for a SaaS product today:

1. Write down a checklist with 5 bullet points in the context of how I imagine the end result would look like of my MVP.

2.  Review if the things I listed are really needed for the first version of MVP and if I'm not giving time to creating the product features that shouldn't even exist the first time. 

Tip: Review it the next day, or a few hours later. So the mind is fresh and that memory is over-writeable.

3. Building the product (as I know how to code): Go with my most comfortable technologies, in my case I've worked with the JavaScript framework (VueJS) my whole life and Tailwind.css and Node.js on the backend

Tip: Do not try to try new technologies. I've tried this many times by trying NextJS / Svelte / .NET  e.t.c or even newer versions of your comfortable technologies. Like Vue 3 instead of 2. 

I know that Vue 2 is old, and Vue 3 is a newer version but am I comfortable with Vue 3? No.

I've done this mistake many times to go with the latest technologies, newer edge technologies. Trying to shoot two shots with one arrow. Achieving to build MVP and also learning new technology.

Only one can be done, either I waste all my time debugging the build processes and my code or build the MVP.

Also what I think may save you time like using an existing design system may end up giving me more pain and just increasing work.

Simply saying: Go with what you already know, what you've already used and are comfortable with. Do not try something new. It's a deep hole you dig for yourself.

If you're not a developer, use Bubble. Don't try to read the pros/cons of all NoCode Tools out there, and compare pricing and stuff. It doesn't matter in long term. Just pick this one, and go ahead. You will always rebuild into code after your idea gets successful or you can always move around later. 

4. Don't try to focus so much on the big big picture, as one of my good friends says "Try to make 1 customer happy first, then 10 customers, then 100, then 1000, then ..." **

For example:** If I'm building an AI photo generator app, it doesn't need to have a forget password page. Don't waste time on that. Your MVP is meant to serve very few people. I'm sure I can manually generate and reset passwords for them when they need it. (Just an example of ignoring things)

What an AI photo generator app needs to have is:

1. An Input to upload photos to model

2. An output that gives you generated photos

If this works, it's enough. Doesn't matter if my login page doesn't handle the edge cases, or my site is not fully responsive on mobile or other cases. **

5. Execute the 5 points list that I wrote one by one, Deploy it. No need for a perfect domain. A subdomain provided by Netlify could also work. Share the link with my first customer. Collect feedback, and build from there.

MVP is done. Thanks

MVP is not taking decisions and setting base for your product for years to come. MVP is fulfilling the need of your first customer and collecting feedback. You can re-build the whole thing in a much better design, technologies after your MVP is successful

Just remember this

I also posted this here:

Top comments (2)

alxgrk profile image
Alexander Girke

Well written! As developers we always try to do the new and cool stuff, but if the goal is to be fast and have a short feedback cycle, what you wrote is very important to keep in mind.

justaashir profile image
Aashir Khan

My first post after a long time