Building a Company Analysis Website

Overview

Ever since I was a senior in high school, I had an interest in investments. I had an economics teacher at this time who posed us a challenge of crafting together a portfolio that outperformed his over the span of 10 weeks or so. The catch was that the students got to choose their assets and allocations (which didn't strictly have to be stocks), but he picked 10 random stocks. At the end of the 10 weeks, if you outperformed his portfolio then you passed the assignment, otherwise you failed it. I recall "buying into" Twitter and Zynga, and ended up passing the assignment, with the main takeaway being that the stock market was something very interesting that I needed to look into more.

To make a long story short, some time had passed and I had read some books, put some money into the market and still had no idea what I was supposed to be looking for or what my strategy was. I spent about 4 years like this. Was I going to trade stocks? Was I going to buy and hold forever? Would I just dollar-cost average into index funds?

It wasn't until I got a 'real' job, my first job out of college that had a higher salary than minimum wage, that I decided I should look into this stuff more. Some time spent on thinking this stuff through could quite literally yield millions of dollars in the future. So, I decided to do some research online and had found out about an author named Phil Town and his approach to value investing. His book Rule #1 Investing was a real eye opener and taught me the basics about what to look for in a company, the different financial metrics I should use to guage how a company is performing, what to look for in management, and the proper mindset to have when buying into a stock. At the end of his book he referenced his website as a place to find stocks that fit his criteria for investment, and this is where the light bulb went off for me.

The Idea

After reading Town's book, I had some other books that I was going to read and take into account when investing. So my thought was, "What if I just built a tool where I can see the methods all in one spot?" I had looked for something similar, but didn't seem to find anything. Not to mention that Town's website is $30 per month and other websites are about the same, so this could save some people some money.

The other books I read were Peter Lynch's One up on Wall Street and Joel Greenblatt's The Little Book That Beats the Market. Both authors are quite renowned investors, and both offer their insights on what to look for when doing analysis.

Execution

Getting Data

Getting data was one of the hardest things about the project. It was one of those things where there was a lot of opportunity cost involved. Most data providers charge an arm and a leg for their product ($5K+ per month) and don't offer free tiers for their API. And you certainly don't want to learn one API, and then have it pulled out from underneath you with huge prices. I eventually stumbled upon a reasonably priced option and the data seemed to check out, so that's what I went with and started to code.

V1

I chose to use Nextjs for this and used subsequently went on to use way more tech than I needed. At the beginning, the tech stack consisted of

Nextjs "front end"
- TailwindCSS for styling
Nodejs server
GraphQL for communication between Nextjs and Nodejs server
Nodejs ETL
Apache Airflow for ETL
Postgresql (only for a few days)
MongoDB

I think one of the problems I was having is that I wanted to learn some new technologies while building because it would "pad my resume", but in the end it was just a waste of time and effort. I even spent time writing my own wrapper for the API in Nodejs, because I thought I would learn the API better if I did that. This was a massive waste of time, and I eventually scrapped it in V2. Watching YouTube videos on how "You should add this to 10x your development" and "This framework does x for y" certainly didn't help with this. The end result of V1 was a lackluster website that needed a couple different servers to run everything, for a site that only I knew about. So not only did the website come out looking pretty poor, but it was costing more than I had wanted it to. This was pretty disappointing because I had spent about 4-6 months getting to this point and I wasn't able to build the vision that I had in my head, and quite frankly, I wouldn't even use the website to do my own analysis. I was also doing all this development after work and on the weekends, so it felt like working 2 jobs, both of which required me to look at a screen and think all day. I was ready for a break and had a trip to Europe coming up, so I finished up any remaining thoughts/development and went on my trip for about 2 weeks.

V2

Another 2 weeks went by before I felt like getting into the project again. I knew that I needed to simplify things and I took on the mindset that I should not be reinventing the wheel. Things should be fast to implement and simple. I would use component libraries, ditch Apache Airflow, and use the preexisting API wrapper to clean up my ETL scripts. I would use MongoDB because I wouldn't need to do any additional formatting when transferring data from the API to my database and it was easy to work with. My tech stack this time around was

Nextjs
- Material UI Component Library
Nodejs ETL Scripts
MongoDB

GraphQL was causing more issues than it was solving, and I was the only person needing to specify what data pages needed, so a REST API would suffice. Nextjs also allows you to write API routes within your project, so there was no need to create a separate express server for connecting my Nextjs project to my database. This removed the need for CORS and was a lot better development experience. Material UI also offered a much better development experience than TailwindCSS, mostly because Material UI offers prebuilt components. There is also a pretty sweet table component that made things super easy to implement. Material UI also uses a similar way of styling when compared to TailwindCSS, so picking that up wasn't hard to do, and in the end it the site came out looking much better.

I also had built out some wireframes of what I wanted and how the site should look and function on mobile and desktop devices. It took about 6-8 weeks to build this version and it came out as I had envisioned. Not only the initial combination of investment criteria, but the additional features as well. One of the features that I use the most is the screener because it's worldwide, and I have built into it fields that no other service offers. I also frequently use the company comparison page, portfolio comparison page, and ticker overview page because that's where I can easily tell how the company has been performing and how they are performing relative to peers and the market.

Getting Users

This has been the hardest part and is still under way. By the time I was done with this project, I had about 1,000 subscribers on YouTube, so I thought that maybe a few of them would be interested in the stock market and this tool. Well, it turns out that was a bad assumption. But I have been able to make use of TikTok, Reddit, and YouTube to get a few different users.

Lessons Learned

Keep it simple. Don't reinvent the wheel, and make use of prebuilt solutions.
More technology probably isn't the answer.
The stock market analysis tools market is pretty saturated.
Invest in great businesses and hold them forever.
100 baggers do exist and future 100 baggers are out there.
Return on Capital is essential.
PE doesn't matter as much as some believe.
People who sell courses/talk about stocks on YouTube are terrible investors, don't take their advice.
Some good people to look into are Terry Smith, Chris Mayer, Warren Buffett, and Charlie Munger.