Boilerplate SaaS tools
Any tools for boilerplate saas apps? I've seen a few with React, but what other stacks would you want to see?
Trying to create something people will use, probably with Vue or Angular.
Thoughts?
Trending on Indie Hackers
Any tools for boilerplate saas apps? I've seen a few with React, but what other stacks would you want to see?
Trying to create something people will use, probably with Vue or Angular.
Thoughts?
I've spent far too much time looking at different SaaS boilerplate options and thinking about the space and I'm not a dev (so never going to compete in this space) so am happy to dump all my thoughts on the topic here publicly. Generally, I think customers in the space approach SaaS boilerplates with a few things in mind, either a) you're non-technical, looking to get a quick start, and assuming you want to start lean with a small team, should seriously consider using a language known for facilitating faster dev timelines and being "sufficient" for small web apps. To me that translates to Ruby on Rails, Laravel, and maybe Django; or b) you're technical, but understand the value of not having to worry about the starter elements that are common to most standard CRUD web apps - in which case you should probably buy a boilerplate that's written in a stack you already know, almost regardless of whether or not it's perfect for your use case - because you need to validate your idea before getting too fancy/perfecting things and you can typically do that much faster/better in a stack you know.
I think the best boilerplate on the market is Bullet Train (https://bullettrain.co/) - I have zero affiliation with them and take my opinion with a big grain of salt, I don't really know any rails. But the development is very active and good portion of the product is open source (no stripe included in the open source product). I also happen to like the default appearance of the interface and landing pages, so that also saves time (some of these boilerplates bring a lot of great back end functionality but look like shit). If you know/can learn rails, then this is great. But if you're hiring devs to do this for you, my guess is that a junior rails dev is more expensive than a junior Laravel dev. The Bullet Train approach more broadly (at least previously) is a rails monolith approach - you don't need crisp because we've got a built-in chat function. Rails gems make this a feasible approach generally. But it's a very different strategry than say divjoy (more on that below). I do think Andrew has seen the light a bit though and has wisely started to defer to tools like circleci for CI/CD etc. - just integrate with really good tools that work well as a starting point for an MVP.
Pros of Laravel relative to Rails and Django are that there are a lot of affordable PHP devs out there, though you could argue it's equally hard to find "good" devs for Laravel as it is for a less popular language like Rails. I don't really know this for sure, but I'm going to take a guess that Laravel plays nicer with wordpress than other things because they share PHP in common. Yes wordpress is uncool these days but it works, and it's fast and open source with loads of content to trouble shoot with and integrations. Don't let perfect be the enemy of good. Laravel also has a host of useful prebuilt tools, most from the Laravel creator - Jetstream, Spark, Nova, Vapor, Forge and Envoyer.
I can't say as much about Django because I'm less familiar, but python is a higher level language than something like Nodejs, and Django is commonly referred to as a "batteries included framework" aka the framework itself is designed to handle most standard SaaS web app functions, so you're less likely to wind up in NPM dependency hell.
I also think these boilerplate SaaS tools compete with no code tools like Bubble, or one I'm looking at closely recently that rarely gets a mention for some reason, weweb.io. As webflow starts to offer better support for their new beta logic feature and memberships, it becomes a more viable way to build a SaaS MVP, and some people are saavy at using Wordpress to build a SaaS MVP.
As for trying to build a profitable business by selling a boilerplate SaaS - when I look at the existing tools in the space I see a few things missing feature-wise that would be of interest - almost none are built to work well with a no-sql database (divjoy is the exception here). While SQL is the more common need, the app I'm working on is more about providing access to a massive database, so my scenario is 10k+ reads for every write to the db, full text search is important, and a few other things that make a nosql db the obviously choice for me. When your customer has fewer good alternatives, you can charge a higher price so in that sense offering a nosql option could be a good idea, but it's also possible that the market for that use case is so small that even the higher price makes for a worse business.
Other features I haven't seen promoted would be - built in connections to Wordpress or Webflow. Again - for an MVP, I want to throw something up quickly and start talking to customers, Webflow and Wordpress make that easy. If you want to lean a more code direction because you think devs won't value the wordpress/webflow connection, then maybe integrate with something like https://shuffle.dev/.
If I wanted to launch a boilerplate SaaS product quickly I'd take the Divjoy approach where the product is essentially a prebuilt set of connections to popular/powerful tools. Defer to auth0, or firebase, or cognito to handle auth. Sentry.io to handle error reporting. Connect with stuff like Hasura, or React Admin, or Firebase, or AWS Amplify etc. - then make it easy to deploy everything via render or heroku or vercel etc. The tools I like that are rarely connected/setup by default are things like Paddle, StaxPayments (cheaper than stripe but more complex to setup), Plaid, MelioPayments, or tools that have generous free tiers like hubspot and crisp. The challenge with this approach though is that your prospective buyer now needs accounts and API keys, secrets, environment variables etc. for all those tools and sites, so you're going to need to either offer a service to get them fully set up (for non-technical folks), and/or extensive documentation to help guide them through it.
All in though - I think most boilerplate SaaS offering suffers from trying to be everything to everyone.
Vertical SaaS is a much better business model for indie devs than "horizontal" approaches. In other words, don't try to take a hubspot approach to being a CRM for everyone, take knock's approach to being a crm for multifamily landlords, or workiz approach to being a crm type thing for field service businesses. There are a lot of ways to add value to the industry down the line and expand your product by taking the veritcal SaaS approach, there's much less competition, and product differentiation becomes a very natural thing for you. In the SaaS boilerplate world this means - don't try to be webflow, be a CMS for a specific use case. I use Blobr.io's product and it's a good example, they are essentially a CMS for an API based product. I'd pay good money for a CMS that is built to help me monitize and control role based access to a large, valuable nosql db (ideally monogdb).
Django is just awesome. Never regret learning it. Though some elitist developers think there's too much implicit "magic" in it, I couldn't care less - it just lets me create what I need extremely quickly, if considered to Flask or fancy FastAPI etc.
Thanks for the write-up! Very interesting read... I am a software dev with a background in javascript (Angular/Vue), AWS back-end using serverless/dynamoDB, C# for back-end (although I have used many of the frameworks and languages you mention).
Some Questions I would have:
Blobr's product is the best example off the top of my head that fits both the vertical saas and CMS description. It makes it easy to create documentation that looks good and goes with your API, does things that other SaaS boilerplates do like handling authentication etc. But then instead of having to be completely agnostic on site layout and design etc. it basically does the opposite where it's very opinionated about how you should design your site, how everything should flow etc. That then allows me to not have to build the site at all really - I just connect it to my API, fill in all the fields it walks me through, connect it with my openapi specification etc. and now I can monetize my api. I suppose you could think of the army of documentation tools this way too - redocly and readme etc are essentially just CMS's for documentation pages and most of us can't imagine managing documentation without them. I haven't used Lodgify but it looks like you could think of it as a CMS for vacation rental property websites.
I wound up on this page because I was looking for something similar for monetizing a large dataset and thus far I can't seem to find it. The marketplaces (aws data exchange, azure data share etc.) are one obvious way to sell the entire dataset, or sell an api. But ideally I'd also have a professional-looking site that would give people a dashboard, a point-and-click ability to query the data, export reports etc (essentially you would just throw in a BI tool like metabase, looker, powerbi, etc. for exploration). You CMS would making it hard or impossible to scrape or otherwise export my entire database or at a minimum let me to know about it if spikes in requests indicated someone was doing that. As well as making it easy for me to design role based access permissions in an admin panel. There's a whole universe of people that sell alternative data to hedge funds and similar financial firms for example, and as I understand it, greater than 50% of them just put up a powerBI dashboard behind a login. For datasets that become useless quickly and must be kept up to date constantly - it doesn't matter as much if someone figures out how to steal your entire dataset from your powerbi dashboard, because it goes stale fast and they still need you. But if you're selling let's say super valuable clean, historical data on different assets for funds to use for backtesting, they could easily steal that and while they'd be missing daily updates to the dataset, they'd still walk away with something really valuable by stealing your data. This is just my example of a CMS I'd like to have that based on my experience could serve a large market of people who want to monetize a dataset that's way too big to put in airtable or sell via gumroad etc. Your CMS would also help by making it easy for them to facilitate and track very large ACH transfer type b2b payments where neither side of the transaction really wants to use Stripe for a $100k transaction kind of thing. I think Blobr takes the correct approach by essentially recommending users design their landing pages in their tool of choice (I do mine in Webflow), and then when you click sign up or login, everything after that is handled by the CMS.
It's both a complex enough use case that I don't want to build it, and a common enough use case that there's no way I'm the first person to have this problem.
Yeah, I could see how that would be incredibly useful... So I am imagining a database of stock prices and movements that are used for backtesting different trading strategies or something along those lines.
Basically, something that allows you to say "Hey, I have this database that I manage and keep up-to-date, if you would like access, you can sign up and pay a certain price depending on the amount of data you pull down"
You can sign up for an enterprise tier that gives you unlimited pulling of the data, and just use some monitoring system that checks for abnormal use or something like that.
The admin could log in and manage subscribers, view their history of use of the data, allow or block certain types of queries, add rules to manage roles (Basically IAM but simpler, if you've ever used AWS)
Exactly. Being able to offer usage-based pricing would be amazing. Tons of large enterprises would like to move from seat-based pricing to usage-based pricing but it can be hard to do.
So for now, people just sell the entire dataset, and customize that a bit further for those who don't want the entire database, but just a slice and want a price quote for that. It should be able to self service that by allowing you to customize a query of the slice you want, see a small preview of what the resulting dataset would look like and a price quote of what hitting the export button would cost you.
In my case, I'm building a massive database of company info. Fields like number of employees, growth of employee count over different time horizons, SEO info, who they've acquired, how much funding they have, or for different businesses, what kind of restaurant they are, whether or not they have a wheelchair accessible entrance, what their tech stack is and how long they've been using each part of their tech stack etc. etc. - a crazy number of fields. Many possible use cases from underwriting, to building a yelp clone, to a wall street analyst who just wants to be able to login and find out how many plumbing businesses there are in the southeastern US, etc. etc.
In most scenarios I would personally want to price it as a "two-part tariff" - at least that's what my econ textbooks called it. A subscription fee, let's say monthly but it could be quarterly or annually, with a usage-based fee on top.
Edit: Shopify is probably the best known example of a CMS for a specific vertical - ecommerce. If you needed a short pitch for what your thing did I'd call it Shopify for monetizing large datasets. The analogy works particularly well in this case because Amazon - aka the marketplace version of the product - pretty much already exists, but plenty of sellers still want/need their own site where they have more control of customer acquisition, higher margins etc.
I have been working in Micro SaaS Ecosystem and here are a few that really picked up momentum.
Hey Divjoy founder here. Thanks for the mention!
Thanks Gabe for joining the conversation. I think you are one of the first guys to build in this space!!
Anycase , good work with Divjoy!! Isn't it supposed to be DevJoy?? 🤔🤔
Yeah it's been a few years! Time flies. Appreciate the kind words. Div as in <div>, but also devjoy was taken 😆
More fun with <div> 👍👍
Something for free please? Why does everything have to be paid :')
Yeah, I see a lot of these feature "sets" and they all seem kind of similar. But a ton of them focus on react - I could see a potential market in some other JS frameworks. The startup world is different than enterprise software, so I am not sure that Angular would be a good idea, but Vue has some potential
I've actually been cleaning up a personal list of SaaS tools to share for others to use. I shared it in a new IH post here, so it won't be buried in the comments:
https://www.indiehackers.com/post/a-detailed-comparison-of-35-saas-boilerplates-a148e12c5d
Here's the direct link to the table: https://t.co/yhZGKZaOvH
This is a helpful table, made it easy to find some examples I hadn't seen that support MongoDB. Thanks for sharing. Bullet Train (my personal fav) may have changed a bit since you made this, but it definitely now has one click deploy (via both render and Heroku) and their "super scaffolding" implementation technically runs via the command line though I'm not sure it fits the technical definition of a CLI in this context.
Edit: Also, jumpstart rails pro also supports paddle now, and has user impersonation.
Thanks, I'll update those once I have time!
Hey Jan - I'd dm you this but looks like you've got direct messages turned off - I also added a column with github stars and # of contributors to your airtable. While they'll be quickly out of date because it's not hooked up via api or anything, it quickly differentiates some of the open source templates from others which I found helpful: https://airtable.com/invite/l?inviteId=inva4HL2aMtV45JSU&inviteToken=d58f42ab660c0b05aab0aa3b1249f0bfed5f2bf5310b57131345bbc1e1d19222&utm_medium=email&utm_source=product_team&utm_content=transactional-alerts
Hey man can I message you? Let me know what platform you feel comfortable with.
For sure - you should be able to dm me here. That will work for now.
Not sure if your just seeing something I am not but I dont see any option to dm on indiehackers haha
Checkout https://boostack.io built with MEVN Stack
I've open sourced my own based on Flask: https://github.com/nuvic/flask_for_startups
I’ve worked with the majority of front end stacks for large and small projects and honestly it doesn’t really matter. Just use one that you feel comfortable with. Personally I like working with Vue 3 using Nuxt but I’ve been using the T3 stack (based on Next) with supabase for my current project and i love it. It has everything you need.
I am trying to put together a product and was looking to get opinions on what currently exists, so as to identify market viability - I also like Vue quite a bit, and i have seen supabase - everything seems so React-centric... What do you think about something akin to supabase, but focused on Vue and AWS or something? Does that sound like something you would consider using?
Ah, I guess I misunderstood the premise of the post (just came across it randomly). Yeah in that case, I agree most of the SaaS tools are mostly react based. Personally I don't use any of the SaaS tools because I usually pick my own toolkit. But I guess there's plenty of opportunity in the Vue space considering I'm noticing a bit of a shift from React to other frameworks/libraries (like Vue is getting more traction). I would also look into newer frameworks/libraries like Svelte, SolidJS and Remix JS.
This also might come in handy: https://2021.stateofjs.com/en-US/libraries
i just grab tailwind css for the ui and coding i think i can handle one problem at a time.
I love this open-source SaaS boilerplate - node, express.js (no db dependency)
https://github.com/makeitcount/saas-express-starter-kit/