The Slashed

The production website of The Slashed is up! GTM is setup for tracking and analytics goodness. It's been A LOT of hours and the milestone of getting the customer facing website up is an exciting one. I will be looking for a handful of beta testers soon to iron out any bugs and there's still a couple of things to polish off from a development perspective, but I can see the light at the end of the tunnel and it's all coming together!

Since the proof of concept in August, I had steadily been building functionality on top of it. As I started working out the concept of a fully fledged sale tracking service, the complexity of the system increased massively. As a solo founder, you have to choose what you spend your time on carefully and make a lot of compromises in terms of functionality, design, technology, and more. It was very much a balance of build only what was necessary for an MVP, but also make it look good.

I don't see many updates that talk about the technical side of building a startup/app so I wanted to cover The Slashed's technical architecture from a high level. Maybe you'll find it useful.

The Slashed, a user story:
As an online shopper, I want to follow the brands I like and be alerted with only the new items that come on sale so I can be the first to know about it.

The Slashed Website: Very simple and straightforward.
NextJS - Server-side rendering, a must for SEO,
React,
React Bootstrap,
AWS Lambda@Edge,
Serverless-NextJS,

The Slashed App -
React,
Redux,
Redux-Saga,
React Router Dom,
Antd Design Library,
Style Components,
Heroku

The Slashed Backend -
Node,
Express,
MariaDB,
Sequelize ORM,
Nodemailer (Mailgun transport),
MJML,
Heroku

Scrappy the crawler -
Node,
Sequelize ORM,
X-ray,
MariaDB,
Serverless,
AWS Lambdas,
AWS SQS,
Docker

Tracking -
Google Tag Manager,
Google Analytics

Like a lot of great ideas, the concept for The Slashed was born out of laziness. I had recently gotten into shopping for more higher quality clothing and therefore more expensive. I ended up following a couple of brands that I knew produced high quality clothing, fit, and suited my fashion taste. With some of these brands however, you do end up paying a lot for the "name" in addition to quality. A lot of items weren't really worth the full price tag to me and were much more reasonable to buy on sale. So I would check multiple sites everyday and browse the sale sections for however many brands I was into at that time. Most of the time I would forget what was there the last time and go through the entire sale catalogue again - times however many sites I online shopped at. As you could imagine, this became a very time consuming and tedious process and eventually realized I could try to automate the process. I had created an app before that would scrape Amazon search listings and figured it would be very similar.

I work as a professional web developer so I thought building a simple proof of concept was going to be pretty straightforward. I wanted something that would go the sale page of a brand on a website, take all the products on that page and put them into a database. This service would run everyday so it should only save the new products it finds. Then it should send me an email with ONLY the new products. It sounded pretty straight forward, but it turned out a lot more complicated than what I hoped. It ended up taking about a month to develop this proof of concept, but the results were positive and encouraged me to move forward with this idea.