Using SaaS starter kits — thoughts?
I noticed there are many SaaS starter kits (i.e. code repo that includes things like user authentication, payment, team management, etc.).
What are your opinions on using them for building your product ideas? What are the pros and cons?
(Disclaimer: My cofounder and I got frustrated about such repetitive tasks and thought it might be an interesting problem to solve. But it seems like there are so many options out there already.)
Trending on Indie Hackers
I've built Parthenon (getparthenon.com) which can be considered a SaaS starter kit, so my opinion is a bit baised. But I think most people should be using one. As you said you're frustrated with the repetitive tasks. This happens constantly. Majority of a SaaS at the start is the same as every other SaaS and that is user sign up, plans, payments, emails, etc You can realistically spend a lot of time building those, trust me, I've spent months building them. If you calculate your time as a cost you'll find buying SaaS framework will be cheaper for you. Especially, since you're probably going to build multiple SaaS platforms trying to find something that works and generates enough revenue.
For me, there are three major players worth paying attention to and which one you pick is probably down to which language/framework you like best.
If you have any questions, let me know, I'll help you out as best as I can.
Edit: A shameless little plug - I've currently got a lifetime 50% off voucher that expires on the 1st of May. IGOTFIRED - story at https://getparthenon.com/blog/fired-to-full-time-indie-hacker
Thanks for the mention Iain :)
I love your voucher's name :D
Fair enough. I do think having something to speed up the initial development is worth paying for. That's why we started working on https://dashibase.com/. It's a slightly different approach to the same problem. :)
There are indeed many start kits out there – I've tried to categorize them in this repo https://github.com/smirnov-am/awesome-saas-boilerplates
This comment was deleted 3 years ago.
Nice. Love learning from IHers on stuff like this.
I've looked into these a bit but have never purchased anything. They seem useful (especially for someone like me who still lacks the necessary skills). Also they could help you get to market much faster if you're a novice (like me).
Whatever gets you moving fastest!
Yep, I agree!
That's the comment my mind was looking for. If a single index.php will do the job, then, why spend time with an overly complicated setup?
Don't do it unless you're prepared to re-do everything from the ground up later.
Yeah, that is definitely a concern. Is it possible to not have to re-do everything next time? I wonder if there are good solutions to that.
Might work for some, and I love Gravity.
But I've never used any because I'm afraid changing them to match my needs would take more time than doing everything from scratch.
Even though they could save time + more robust code, I don't use them because it removes the part I like the most: building from scratch.
I can see the appeal behind starter kits and they are certainly useful for some people.
However, I personally wouldn't use them. I have pretty strong opinions about how to do things. So I'd change half of the things in the kit anyway. And that's often hard to do or time consuming.
Plus, after you have built your first SaaS with a stack you're happy with you can use it as a base for new projects.
Or put it up for sale as another boilerplate :P
Big brain thinking :D
That is fair. I imagine those starter kits are targeted at people who are unlike you haha. :)
In general, I think they are a great idea.
Any system which can speed up development, helping you get to market are worth at least taking a look at.
Whilst I've not used a 'SaaS starter kit' as such - I use Laravel to develop my web apps which itself is a framework that really speeds up my workflow.
This time around I also invested in Laravel Spark which is a first-party add-on that scaffolds all of the payment process. This was a no brainer as it would have take seevral days to do this manually.
There are no doubt lots of solutions on the market but they can not be everything to everyone, and nor should they be.
If you find your niche and develop a robust system to save a bunch of developers time ad money, then it is worth exploring.
Thanks for the encouragement, Andy!
Yeah, from a marketing standpoint, it is not a great idea to be everything for everyone, especially as we are starting out.
I'd 100% use them if properly documented.
Fair! How do you feel about having to understand all the code though?
If it's easy to understand code & well documented with quick & responsive support from the provider, then I'm:
Documentation is key indeed
I'm working on building a CLI to launch SaaS projects, sort of like an "npx create-react-app" but for SaaS specific projects.
So I may be slightly biased, but I think if you want to ship an MVP ASAP and spend less time developing, then it's a no-brainer.
I think there is a lot of space for such tools. :)
Where can I check it out?
Thanks! Will send one when it's ready, should have a link by the end of this week/early nest week!
Cool let me know too...
Do you have a link? This sounds interesting
Thanks! Will send one when it's ready, should have a link by the end of this week/early nest week!
What are wrong with other solutions?
I think my hesitation is that I need to spend a lot of time understanding the code and building from scratch might be easier (though still hard). I also don't know how to choose between all the options and am hesitant to spend too much time on one and have to switch later.
Why do you need to understand all the code? Most of the functionality you will use as is, or maybe with some minor tweaks. Like using any external package, you don't need to understand how it all works, just how to interact with it.
Maybe not all the code but some. E.g. where to hook up the database, add variables, etc. But yep, I think we are thinking about the same thing. Maybe because I'm not a strong developer, so I take longer to understand some code.
A good boilerplate will use standard conventions for these things, so it should be relatively intuitive to figure out how to do it. In https://usegravity.app I have a setup wizard that handles all this, so users don't even need to think about it. They just add the db creds and API keys into a UI and then the app configures everything for them automatically.
Oh, that's awesome! I think that UI makes the difference.
Understandable. Also I read it wrong and thought you wanted to start one.
I personally think there is a gap in the market for a Supabase/React/Tailwind SaaS Boilerplate.
However there isn't a default Tailwind CSS and React framework, until that happens it ain't worth building.
Not sure about Tailwind CSS frameworks. For React, I guess NextJS is the most popular framework?
Also, we are thinking of solving it in a different way because we are not sure boilerplate code is the best solution. We have a prototype at https://dashibase.com/ if you are curious!
https://divjoy.com/ Is react and Supabase. And soon tailwind i hope. Waiting for tailwind before i buy.
Is divjoy a SaaS boiler plate? Looks like landing pages.
Did you try the builder or look around? On the front page you can choose auth and payment. It says:
Divjoy gives you all the UI you need. Landing page, authentication, pricing, forms, account settings, payments, billing, and dashboard w/ live data. All responsive and perfectly integrated with your chosen stack .
I have always been confused by their home page. It just has landing pages.
Is there any screenshots of example account settings and other dashboard parts?
Here’s s walkthrough:
https://youtu.be/cUSEWCqe6Ig
And on the front page you can choose a build. Then it takes you here:
https://divjoy.com/build
Most of the starter kit address requirements at surface level. They provide modules like user management, billing, teams etc.
But to run a SaaS business, it's not enough. It also needs
Hence, most of the business who use stater kit eventually end up rebuilding software from scratch or heavily making changes to the starter kit.
I have been at two companies that rebuilt the software from scratch despite building it from scratch initially. Maybe it's inevitable? 😅
Also, I guess it depends on the scale of the business. Not every app needs to become a huge business.
My boilerplate has almost all of these things, apart from queuing because it is not often required and is easy to setup with external packages when needed.
Do you mind let me know which boilerplate did you use ?
Sure, the one I built :)
https://usegravity.app
Gravity is good. I recently built my own stack
I have implemented all the above features I shared above plus few more...
One Good part of Quasar Framework is you can build multi-platform app using single code base. You can even build a browser extensions.
Good you brought this up. Has anyone used https://jumpstartrails.com/ ?
Sorry, I have not tried it before! Perhaps you could check reddit for reviews?
I worked on dozens of SaaS products and what I can tell you is that there isn't a one-size-fits-all reply to this. It really depends if your specific needs happen to match with the kit.
Also, you need to think if the structure is going to be scalable long term as you start adding features or if the product makes a U turn in some areas.
That's fair.
I would love to learn from you. We have been working on https://dashibase.com/, a no-code tool for building an app dashboard on top of Supabase. From your experience, what types of businesses or teams do you think would have such a need?
That's an interesting concept.
I will start by saying that I'm a UI/UX designer, so take my development/coding with a pinch of salt, but I think that mostly people looking to kickstart a project or anyone who is in early stage with a setup that allows for flexibility would be in your target. I think the more they grow into their structure, the less likely they will be to make that U turn. I'd try and market on places where new potential founders are (people who have the new startup ideas), and even be visible on places like Betalist, Product Hunt and try to get in the incubators network by developing relationships etc.
That's awesome. Thanks for the suggestions, Pierluigi!
SaaS starter kits and Design figma files are the new Todo apps. I doubt the world needs that many of them.
It depends on the objectives. For building a side project as a hobby, then building it from scratch might be the way to go to save money.
If it's a commercial application, then sinking months into building something that can be purchased off the shelf for a few hundred dollars is not a smart business decision. That time (and money) can be better spent on building features that customers will pay for.
Why build everything from scratch? Composability is power
Right, but it is also a matter of how you assemble existing solutions to meet your demands/requirements.
Sometimes, you have to build (at least parts) from scratch to customise to your needs.
Of course, but a boilerplate gives a huge head start. It's much easier to make a few adjustments to an authentication system than to build an authentication system from scratch.
I think it really depends on what you want to build, the amount of time you have and your experience.
For me, I decided to build my own because I have several SaaS ideas and wanted a template to fit with my prefered stack that would save me loads of time.
If you're looking for one built with AWS I have a beta version available now.
Here's my landing page: https://laaso.dev
Personally, I would prefer building one for myself over buying a starter kit and using that, because you can build it to your needs.
For me the repetitive tasks are pretty quick to do when re-using code from older projects.
If you’re going to spend time reading to other programmers code, might as well learn a new framework in the process. So choose one that has framework you’d like. I have my own SaaS starter kits in 4 frameworks, I’d 100% choose the Remix one, I could give you v0.0.1 for you to test 😁 (right now it’s at 0.2.4).
They are a great idea, but they are also very personal.
That's why I started to build https://businessclasskit.com/ for Rails. The existing ones didn't cut it for me.
I'll offer it for others, but likely won't have many customers due to the fact it's opinionated (it's for me after all!).
What I don't like about the ones existing:
Yeah everyone has their preference for the stack, and perhaps that’s why there are many different starter kit.
We are thinking of abstracting some of that but then the target audience might not be developers who like coding all these.
I find that it’s very easy to diverge from a kit—different framework, different database, different deployment, different payment provider, that I haven’t been able to get my money’s worth out of a kit. I end up re-implementing many features anyway.
That's fair. I think there are some kits that let you pick what framework, database, etc. you want to work with. Have you tried those before?
I haven’t seen too many options like that. Though, I doubt they would have worked with my current setup.
It's a waste of money. If you know how to code you can reproduce what many of these starter kits have created in a couple days.
This is an opinion that is often shared about many things. Famously, Dropbox received the exact same style of response. Where someone said they could easily reproduce the same results using rsync.
The major players (getparthenon.com, saaspegasus.com, and https://usegravity.app) have all had months of development put into them. Therefore, realistically, it would take months to reproduce what they bring to the table. Underestimating how long a task takes is so common in software development I've literally been asked in interviews why tasks always take longer than estimated. Hell, for most developers just writing the automated tests for one part of the system would take them days.
I view it as a quick repackaging of open source software. I'd much rather donate money to the actual open source software creators and maintainers. Being charitable, I think the value that these starter kits add is knowledge of all the different things required to run a SAAS but the problem is, if you are knowledgable enough to do things like run and setup an AB testing framework then you are knowledgable enough to do it without a starter kit and if you aren't knowledgeable enough to know what AB testing is, a starter kit will be a complete waste of money.
I'd like to see actual SAAS products that are being started with these starter kits. If they really brought in a ton of value you'd see many companies being started with them.
The repackaging of open source software ones are the low-grade ones. Not any of the major players where they've got lots of code written by them
The value is that you don't have to spend time doing basic stuff like user sign up, forgot password, invite user, managing team, managing plans and manage limits, integrating with stripe, etc. And getparthenon.com for example even has A/B testing, and the ability to have tenants on their own database.. It Even if you know how to do all that which most customers do, it's still time spent working on stuff that is not the core product, it's time spent one stuff that doesn't make you special, it just gets you to the entry-level. if you value your time and say charge even just €25 if you spend 40 hours, your standard work week you would have spent more money than it would cost to get a bootstrap, have less features and get no support and have to fix your own bugs.
Simply the value that you save time and money.
This is simply not true. For example, authentication is one of the most complicated things and requires a lot of time and effort to make it properly, test well, remove bugs etc.
I never suggested building your own authentication. There are a million services that exist that solve authentication for you
see:
Google Identity Platform
AWS Cognito
Auth0
etc
Implementing one of those services is easy.
I've never talked about implementing authentication from scratch.
The truth is even if you use "out-of-box" solutions it always takes time to adopt it and it's not about a couple of days.
I'm curious to know if starter kits even solve that problem. I think with a starter kit you'd still need to spend time learning how the chosen auth provider works, same with stripe, etc.
You shouldn't. Check out our boilerplate https://www.saasforge.dev we already implemented the whole authentication logic and it can be adjusted. Here is the documentation on authentication https://www.saasforge.dev/docs/modules/auth. All you need to do is to provide some environment variables. It may takes some time to learn which to provide, but not too much.
Exactly, building and testing a full featured authentication system takes at least a month to do properly.