Ask HN: What are your biggest roadblocks when building a new project/SaaS?
I find that auth is always a personal pain point whenever I start a new side project. I'm curious what other people find as their common stumbling points.
The Process is usually
1 - Think of idea
2 - Research idea
3 - Continue researching the idea for several months, promising myself that the more information I have, the better equipped I'll be to tackle the project
4 - Idea goes from "Simple idea X" to "This will be the new Google"
5 - Get so wrapped up in the previous stage that I lose track of the simple original idea.
6 - Get overwhelmed
7 - Begin to think "Well, X does this idea better anyway, why even bother?"
8 - Sulk
9 - Forget idea
10 GOTO 1
Most of the ideas seem to be: find niche, validate with a few customers, then just push for a few years till it works at scale. Skipping the months of research and everything else after step 2.
I built an auth system seven or eight years ago that I regularly evolve and reuse. That goes into place on the first day, with usually minor adjustments per project and it just works. Before that auth was always an annoying chore.
I do the same thing and do it for the entire “core”. Templating, session management, database tie in (all for a stateless, cloud app).
I improve it every project and it’s easy to start from there than scratch. I don’t like using available base code due to not needing 90% of features and it always feels like bloat.
Its usually new tech or in a new domain so I do all the research and get the workflow from beginning to end working.
The problem is, this milestone is usually 85% of a prototype and 20% of a product. And at this point all the fun and motivation is gone and "launching" was never so interesting to me as "getting this finally to work".
So usually I commit my code at this point, push it and go do something else. :)
Honestly it only ever worked once for me and that was when I showed the prototype to an audience and they kept me motivated to keep going but I have a feeling this is not the best way to go about it.
To me this seems like the way to go - build a prototype and get it out there as soon as you can. The feedback will keep you motivated, show you if the product is good/useful/needed, and guide you on what direction to go next.
On Node.js projects, using passport isn't quite as easy, so that is a bit of a pain point.
My biggest stumbling point is probably the UI - it's easier for me to work with something that "looks" like a finished project, than it is to work with plain basic Bootstrap 4.
1. what is the motivation - a pain killer or a vitamin
2. Decide - Is this to add income to your regular job
3. Decide - If not 2, think of cost to run it
4. Decide - How do you want to manage this, when you wake up next day with another cash cow SAAS idea
5. clarity - time you want to spend on this EVERY DAY, to not regret later
6. Decide - What is that only feature you want to release in the MVP
7. Set - timelines, STOP pressing yourself into the stress; easy goals and relaxed timelines
9. Launch - no better place than HN to release MVP
Often, people ignore supplements, cannot ignore the pain killer.
Eg: Yelp (pain killer) solved the pain by putting all restaurants together. I built a (vitamin) plugin (lemony - chrome extension), to display restaurant's FDA ratings - while you can't completely ignore FDA ratings but easy to overlook.
Building a quick MVP and running closed alpha/beta tests helps me push through it though. Getting feedback on your project and feeling some kind of progress is a great cure.
1. db model
2. orm/client mapping
3. some sort of ui that can crud the above