Feedback on my product's direction, please?
Hi awesome Indie Hackers,
I've recently launched the new website for my SaaS, stackmate.io and as I'm now splitting time between content creation & new features, I had this thought:
What if the product only manages cloud infrastructure? For example just deploy the AWS resources a developer would need and be done with it. The developer would then deploy their apps their own way.
The reasoning behind this is that my scope would be smaller, so I would be able to provide a more focused and robust solution. However, I see all other deployment startups going full circle.
What do you think? Thanks in advance.
PS. If you're interested in chatting with me, feel free to schedule a call https://savvycal.com/falexandrou/chat
Trending on Indie Hackers
Hi,
I'm a AWS developer and our team uses cloudformation and also some legacy terraform to manage all our infrastructure as code. I guess your target market must be developers not familiar with cloudformation, terraform, etc.
How exactly does your product help developers with OR without infra skills to manage AWS infra? To me this isn't clear on your landing page
And why wouldn't they just create their infrastructure via the AWS console. How does your product save a developer time?
Thanks
Hi there,
Thank you for your valuable feedback. Yes, the primary audience is developers that are not familiar or don't want to bother with IaC.
To be completely honest, I have this feeling as well, that the landing page is not clearly describing what you can do with this tool.
To answer your question, let's take a Rails project for example: We scan the project and automatically populate the services that are found (eg. databases, s3 buckets, CDN, caches etc), the software dependencies (eg. puma, sidekiq), configuration files, environment variables.
Once you hit the "deploy" button, you'd get a VPC, RDS databases, EC2 instances etc, provisioned with the dependencies and configuration and every time you hit git push, your code will be deployed automatically.
So do you think that my messaging is wrong or perhaps the product is too opinionated and developers might need more flexibility? (eg. use their own ways of code delivery on infrastructure that would be deployed by my tool, as most of the people I demoed this to, focused on the "no-code terraform" part)
First of all, congrats on getting this far. I know I jumped straight in with the feedback. I would need to try it out to feedback in more detail
Well I agree with you about the landing page not describing the service. Your example helped to clarify this. To fix this I would add a "how to section". Just add some numbered steps like this :
Keep these points high level so it's easy to read and add some images to help.
Yes, I would have cut my scope to just include the infrastructure and make a decision on delivery much later on.
I would use the user subscription figures to guide you. What have these been like? But it's only when a customer pays you know they really like it
Some thoughts on this. This sounds very similar to elastic beanstalk. Can I use docker with your service?
Did you build the service and landing page yourself?
Hope this helps
Thank you for your kind words. I 100% agree that the landing page is not very helpful.
The subscription figures are pretty encouraging but I think if the website was better, they would be as well. For example, I saw that only when I was giving a demo and was answering questions I got the "yes, I would pay for this" answer, but not all people would be interested in a demo. My hunch is that the majority wants to onboard themselves and try to figure out their way around the product.
Docker integration is what I'm currently working on and this is where I started having questions on what needs improvement, due to recent advancements in Docker tooling that enable you to take over the deployment part, but still having to create the underlying infrastructure (eg. databases, cache etc).
I built both the service and the landing page myself (to be fair I used a ready-made theme). Obviously, coding is what I'm better at and I now need to focus on the other (very important) parts as well :)
You did help and I'm grateful for that. I'm already trying to figure out how to better position what I have
Curious who your target market is - what type of user? If the user is technical, I would expect them to use CloudFormation or Terraform. If the user is not technical, I'm not sure they should deploy directly to AWS, perhaps a PaaS (e.g. Heroku or Render) is a better fit. Do you have a clear definition of who your ideal user is?
Hello, my ideal users are freelance developers that want to focus on development instead of Ops. I had 30 customer interviews with prospect clients and told me that deploying to AWS for them is a big hassle: they would either hire someone to manage their infrastructure (eg. via Terraform / Ansible), or they would spend quite some time in the AWS console. Let's say the ideal user is the busy freelancer who just wants to deploy to AWS.
Happy to hear your thoughts if you think this needs refinement
Good to hear you did customer interviews. In my opinion, no freelance developer should be deploying directly to AWS. If asked for my recommendation, I would (and do) send them to Heroku. It's hard to argue against it for small scale apps, early prototypes, etc.
Now, considering Heroku has been on the decline for a while, perhaps it is falling out of favor. But the point is more that as an app developer, you shouldn't be focused on infrastructure at all if you can avoid it.
What Heroku showed us was that you could have a dead simple deployment process that was literally just
git push heroku main. The beauty was in the simplicity of the developer experience.I would bet that given the choice, freelance developers don't want to be using a cloud infrastructure management interface at all - be it AWS, yours, or any other. They just want a simple way to deploy their app and get back to working on development.
Do you think your platform serves a need that isn't met by current PaaS providers?
Well, a couple of things here: freelance developers are deploying directly to AWS and why shouldn't they? For example, I had a couple of clients that hired me to manage their AWS infrastructure, so that their contractors would deploy there. I also found through customer interviews that sometimes customers request their projects to be deployed in AWS.
Yes, the quest for a "better Heroku" is a road that leads to nowhere however, there are people who find it too exhausting to properly set up an EC2 instance, security groups, databases, manage secrets, configuration files, automate deployments and that's what this tool is trying to achieve. We also don't lock people in by reselling AWS or cap its prices so it's not a PaaS, it's merely a provisioning tool that abstracts an enormous amount of complexity.
I also agree that nothing beats a dead-simple experience and that truly is my top priority, hopefully I'll take all the necessary steps towards that direction.
Sounds good. Good luck!
Thank you!