Estimating My Project's Costs
Hi Everyone,
I am an entrepreneur that doesn't have a lot of coding background. I am looking for opinions on a reasonable cost to build out an AWS hosted SaaS product that lets a customer input a few keywords, and then scans (week or daily interval) approximately 50 websites for any new content containing the selected keywords.
I would like to have realistic expectations on the cost to build an MVP like this before talking to any developers.
Thank you!
I have not tried this but there is an app for estimating the app cost: https://www.producthunt.com/posts/estimapp-2-0
Hey thanks for the link. That would be a useful tool for freelancers! The app came back at just under $18,000 based on the info I entered. I haven't hired enough developers to know if that is realistic or not!
What you said can be done in five days top, including frontend, backend coding, server setup, basic server security settings, load balancer with a few API end points at various server locations because you wouldn't want this to go down. There are many cron job pre-packaged choices you can use to achieve this.
$18,000 definitely is a lot of money got tossed into an idea that hasn't been validated with customers yet!
At a rough guess, for a sketchy MVP, a competent developer should be able to put that together pretty quickly (like in real terms it would take them, say, 10-14 days).
But, freelance, I doubt you'll get many quotes at less than a month of effort x ($LOCAL_DEV_DAY_RATE)
Bear in mind that this is MVP. If you achieve product market fit and get some real traffic, scaling the backend scanning jobs becomes the technical challenge. This will cost you lots more in dev time and much more AWS cost (extra instances at $$$)
Based on your explanation, the MVP does seem simple. The developer would need to be a full stack dev, that is familiar with both front and backend. It sounds like the core component will need to be a scraper. There is many open source ones out there, so they will not need to reinvent the wheel. They will also need some exp with background processing because these scrapers will essentially need to run as async job on your server.
The MVP is simple, but I have a feeling based on experience with the developer, the implementation could be done well or badly. Since the architecture is important in deciding the scale for later on.
I am based in Toronto, and I think for an MVP for this project, I would estimate $3000 to $4000 CAD (this includes setup with AWS and deploying etc)
I would say your AWS costs, can be kept low for the first year, due to their free tier for a year. You can keep just one server for the MVP stage and put a few background workers with the same web app. Then expand out when the need arises.
I am unclear if you are asking about total price for the MVP including infrastructure and developer or just what your monthly spend could be on AWS?
The question was meant to generate cost expectations for developing the MVP and setting it up on AWS, not the monthly AWS spend.
Gotcha. I would say an MVP for something like this based on your description would be doable by a decent developer at around 2-3 weeks so figure their going rate at $100-150/hour (which is on the very low end for some) at 5-6 hours a day, you are looking at somewhere around $18,000 for development.
I suspect that AWS cost will be pretty low for you if architected correctly by your developer as you should be able to capitalize on your free tier for a year to get things off the ground. Ball park would be under $100 a month for your initial MVP in AWS cost.
This is my ball park and it could certainly fluctuate depending on how complex your project is and the developer you bring in. I hope this was helpful at least...
There are a few variables you still need to decide on. Will you need a brand designed? Will you need help validating the idea? Do you need payments infrastructure built in or will it be a free service? How will people log in? Price will also vary a lot depending on the experience level and location of the developers.
Thanks for the input! I will do the branding/validation. I will need to integrate a subscription billing service and user accounts.
It's a good thing that you do the branding/validation on your own because it's a long term battle and no one knows any marketing tricks will work for sure until after they have happened.
Thanks for the responses!