Have an app? Have a blog? Just want to get alerted when someone online mentions your name? Now you can.
Not too long ago I had been working with the idea of flywheels and I built a bunch of iOS apps but I wanted to start a project that was a web app. So that’s what I did. I’ll explain all about it below.
This past Wednesday (Jan 24, 2024) I did a soft launch of a new app called GrowStack. I posted it to a few subreddits, hackernews, and linkedin. The response I got was decent. Slow start but that’s how all things are at first. I’m going to just keep hammering away at promotion because I really believe this tool will help people promote their own services and apps. The webapp is called GrowStack and currently what it allows someone to do is enter keywords that they would like to be alerted on whenever they are mentioned online. It’s completely free and I did my best to make it as easy as possible for folks to sign up. Check it out at https://growstack.app if you’re interested.
Right now I’m scraping websites like Reddit, Hacker News, and IndieHackers but I also opened a poll for the next website folks would like me to scrape. Building a scraper doesn’t take too long, a few hours if the site has an available feed (let me know if you’d like a post about how I’m doing it; I’m happy to share). Probably a lot longer if I had to do it using HTML scraping. If that was the case, I might pass on that site and focus on lower hanging fruit. There is also a tool called diffbot that is really good at online scraping but the price is steep so I don’t think I would go that route unless the service had a higher impact.
Flywheels And The Accidental WebApp
A flywheel is the idea of the output of a system leading to increased input in a system. The idea behind GrowStack was to have a flywheel for mobile apps with three stages (Attract, Engage, Delight). The majority of my focus on the app was in the engage portion of the system. I built a user tracking system inside of GrowStack with a client side component for iOS (similar to mixpanel) but when got a spark of inspiration while I was at the gym one Wednesday afternoon, I stopped development on that and shifted to the Attract stage.
My family and I were partying on new years eve and the thought hit me while having fun that it would be awesome for the attract stage of the app to have a way for users to get the most up to date posts from reddit and online sites when their niche was mentioned so that they could be the first person to drop a link to their service or build goodwill with potential users. But that thought lasted just a few moments and I then forgot about it mostly because I didn’t think I would be able to build a system like that. There’s an enterprise type of system that does this for large corps call social listening and those system analyze and extract sentiment and all kinds of data from social network posts. This was way out of my league…
So I forgot about the idea and just kept hammering away at the engage portion of the project.
The Impact For Others
For GrowStack there was the thoughts that I could release a light weight free tool that would allow people to enter their keywords and just get a post whenever a new post on sites contained their keyword. I got really excited at the potential that the idea could have in helping others and I decided to build it. What’s funny is that it was just going to be a small tool that would be a lead magnet for the main application but as I thought about it more I realize that it was the perfect product for the attract portion of GrowStack. Exciting!
Credit Where Credit Is Due
I have to give a lot of credit to Lewis Van Winkle, who created F5Bot. The service he built does basically the same thing as GrowStack and was a tiny side project that he built that grew into sending about 10,000 emails per day. Pretty successful in my opinion. I even reached out to him to see if I could buy the app off of him but we decided not to do it. But in the mean time I made a great friend (he’s a really cool guy) and also he wrote a blog post that explained a little bit of the internals of the system and that was just the inspiration I needed to do it myself.
Tech Stack
At the time of this writing, I’m a full-time iOS engineer but I’ve always wanted to build something using ruby on rails even though people think it’s going out of style. The next best option I had was to use Laravel and it was actually the more sensible option since I have a background in PHP development and I also had a license to Laravel Spark in my back pocket. Side note, this is always the toughest part of a new project for me – deciding on the tech stack. Even harder is finding a good name that is available but I’ll talk about that another time.
Anyway, I decided to just bite the bullet and write the thing in RoR even when I have just about 0 experience in the framework. I did write a few blog posts and YouTube videos about building apps with rails but folks were learning along with me in those posts. With the help of Copilot and ChatGPT I was able to get something going. I was also using the Jumpstart Rails project as my starting place and the guys over at GoRails did a fantastic job there at build a great Saas starter. Without jumpstart the app would have never gotten started because of all of the nonsense work it would take just to start build features jumpstart already had it baked in.
This was the tech stack that I ended up using
- Ruby on rails backend
- Stimulus and vanilla javascript on the frontend along with standard ruby erb templates
- Postmark for email
- Postgres DB for storage
- Google analytics for analytics
- HubSpot for marketing emails
- DigitalOcean Droplet for hosting
- LetsEncrypt for free SSL
It’s a pretty lean stack and I’m happy with where it’s gotten the project.
Moving forward I have plans to use VueJS or React on the frontend to build out more complex components as need. I have more experience in VueJS but work uses React so knowing more React could make me more valuable at my dayjob.
Where From Here
So the app had a soft launch and a few users signed up. I’m going to continue to promote the app on reddit and also I am using GrowStack to get alerted when keyworks related to the app are mentioned (I’m eating my own dogfood!).
I have a lot of great features in mind for the keyword searcher. I do plan to have premium features released in the future but it’s my core belief in giving before receiving so regardless of what does get built as a paid feature, the core of the tool is remain free.
I’ll be investing a few hours each day outside of work and other commitments when the kids are sleeping to add more features and address the feedback I’m getting.
Overall I just want to say thank you to the people who did try it and are getting value out of it. I hope it continues to serve you and you see great benefits of being the first to know when your keywords are mentioned online!
