Ask HN: Best practices to kick start a new SaaS side project?

[tootall]

Hi

I was revisiting my personal list of "passive income" ideas, and I found a problem that I could solve by creating a very small SaaS product.

In my full time job I'm a core backend developer and devop so I'm pretty familiar with that part of the stack.

The business logic of this idea is well defined in my mind and I absolutely would just love to start writing the classes that implement the feature (I am proficient in both Python and Java), but I get too overwhelmed by the other components that sound very little appealing to me but are extremely important for a user facing product:

1) A very simple (yet good looking) website that includes a front page and user section (settings page and a few other pages)

2) Manage user accounts: custom user/password signup, google/facebook login, forgot password, ...

3) Manage payments: I'd like to start from the beginning with a PayPal button to enable users to pay something like 0.99$/mo after 1 month free trial

I don't want to rely on too many expensive external services because I want the service to stay up and running essentially for free even if I don't get a single user. I'm planning on using either Heroku or GAE considering their reasonable free tiers, but I might not exclude a 5$ DO droplet as well. I'm familiar with all those 3 platforms so any of those would work ok.

Essentially, with my current skillset, I think I could complete the core of the feature in a weekend but then it would take me many weekends to address the non-core features mentioned above. Do you have any suggestion?

[michaelbuckbee]

Before you write any code or set anything up, try and find a dozen people with this problem and see if they'd be willing to commit to trying out your solution once you make it.

Also, the biggest expense with a side project that you're trying to turn into something more is your time, spend money on whatever helps you ship the quickest. With that in mind, with respect to your points:

1. Buy a theme off of Themeforest for your site - search for "admin" themes as these will have lots of nice predone pages with charts, user sections, etc.

2. You mentioned Heroku - you could use a User management app like https://addons.heroku.com/auth0 - or just implement whichever one system was the easiest. Again, whatever you can push into actual user hands the fastest.

3. For payments to get started Paypal is fine, but was 99 cents per month an actual price you'd charge? That seems crazy low for anything.

Good luck with your project.

[tootall]

Thank you very much for your answer! To address your points:

- I'm not doing a preliminary marketing research because, in the worst case, me being the only user of the service would be enough, since it's a real problem that I'm facing

- Pricing: I was totally guessing, it would not be an application the user would have to interact a lot with and there will just be a fairly low amount of background jobs, so I don't see the pricing go much higher than that, but then again, me being the only user of the service and using it as a portfolio project would be already a great goal

[jf22]

Be realistic with yourself.

Are you doing this for income or to scratch an itch?

Charging 0.99 for something is WAY to low to even break even once you factor in tax, operating costs, marketing costs and support.

[tootall]

Thank you for your suggestion.

As I said in the previous comment, it doesn't really matter, since I want to solve the problem primarily for me by minimizing the costs, so my current problem is adopting the best technology, marketing is orthogonal for the time being.

[sejje]

The difference in .99 and 4.99 is--non-existent, really, for the vast majority of end-users.

If it's useful enough for you to spend all your time building it, and I'm in the same position as you, it's probably worth $20/month, but it's most definitely worth $5/month.

Not trying to harp on this part of the thread, but it's the overwhelming theme of your responses for a reason.

[redtexture]

A point of view on price-setting, and how a dollar a month is undervaluing your idea and your effort.

Let's say that it takes from start to "finish" (knowing that nothing is finished), a modest 50 hours of your time to set this up, and that you hypothetically value your time at $100 an hour, and you have concluded this solves a problem for you over the next several years that makes this effort and mode of valuation sensible. Fill in your own numbers for what you situation is. So, a hypothetical $5,000 investment of effort (or whatever value you come up with) will in your assessment be returned by the value generated by the SAAS project.

Turning this around, toward having others support the project might look like this:

So, if there were 10 or 100 others that also could benefit from the project, and you wanted a two-year payback, plus nearly zero income to pay for operational headaches, processing fees, etc., for 10 clients, you're looking at above $25 dollars a month (10 clients x 24 months X $25 = $5,400 gross income) At 100 users, that works out to above $2.50 a month, for $5,400 gross income).

[funkylexoo]

If you're planning to build a little SaaS, you may want to check out the book "Lean Analytics" It will help you structure and priorities the different activities when growing your SaaS. Note: I have no relation with the authors or publisher. It just happens to be one of the few BS-free business books out there.

[genbit]

I'm working on a software for creating Web UI for SaaS services. If you interesting, send me email, I can help with frontend.