TruckRentalPrices.com

#1) Start with the business model. How will it make money? I've had luck in the past as a publisher earning a living off of ad revenue from Adsense and the like. In this case, I wanted to add another site to my portfolio as another revenue stream for ads. The model also lends well to affiliate partnerships, sometime in the future.

#2) Get the data. Actually, start with a data model to be precise. The site compares the prices of truck rental companies, and there are many across the globe. To keep it simple, the MVP only covers the US (sorry int'l folks), and the six leading companies that have the most market share. If you are familiar with SQL or even a spreadsheet, a good idea is to create a list of columns that will store the attributes. In my case, it was the location data such as coordinates , addresses, and of course, rates. Rates was the most challenging because of the nature of the industry. Scraping data was the most time consuming process, mostly because these corporate sites suck, and much of the data is not structured properly. I stored the data in a postgres SQL table.

#3) Build an API. Once I had the data in a data store, an API exposing is critical for a web app. I used the Express.js framework built on top of AWS Lambda@Edge using the Serverless Framework. This architecture is super fast, and you only pay if you go beyond the free tier. No users, no bill!

#4) Build the front end. In my case, I had experience with Next.js, which is good for making page fast and server side rendered for SEO. I used the Google Material UI library because my design skills are not so great, and the prebuilt components look nice and Googley. The front end fetches the data from the API to build the pages.

#5) Deploy. The makers of Next.js make it easy to deploy to their cloud (Vercel) and point a domain name to it. Vercel has a free tier and then jumps to $20 a month beyond that.

#6) Register a name. I'm old school and I like to stick with .com extensions. Every time I see a keyboard that has a built in ".com" button, I think to myself, there's a good reason to use them. I create lists of keywords around my niche, mix and match them together, and use a WHOIS API to check for availability. In this case, TruckRentalPrices.com was available, and it accurately describes the product. It might be good for SEO to have the keywords in there, and the phrase does get a few thousand monthly searches. Point the new domain name to Vercel, and you are up and running.

#7) Setup Google Analytics, Google Search Console. It's free, and you need to see what's going on behind the scenes.

#8) Get the word out. This is where I am today. I'm a little bit stuck here since its not the sexiest product, and the niche is quite broad and competitive. Still trying to figure this out. Ideas in the comments are always welcome :)

Now that it's out, I'll wait a little bit to get feedback and see if there is any traction before investing in more features, data, etc. Most of the sites I build never get past this step and lay in the abandoned graveyard. In the old days, I had to shut them down because hosting was expensive. Nowadays, I'm only stuck with the $12 year domain name to keep it running, and maybe a few bucks in fees from AWS. The great thing about the abandoned graveyard is that some of the sites do get a hundreds or thousands of visitors a month without any effort. With ads in place, some will do $50 or $100 a month or so in revenue. This is generally my worst case scenario. If it goes no where, at least my internet bill or monthly coffee expenditure for the month is covered!