# Github Copilot helped us cut down 50-75% time in our e-commerce business

> Source: <https://dev.to/heyitsmarcucu/github-copilot-helped-us-cut-down-50-75-time-in-our-e-commerce-business-10l5>
> Published: 2026-05-26 08:08:46+00:00

*This is a submission for the GitHub Finish-Up-A-Thon Challenge*

A couple of years ago, I went back home *after a decade of being a full-fledged web developer* and started helping out in a family business. Of course, obviously I had to learn the ropes from the production floor up till office work, since it is an e-commerce centric business.

Then I noticed something, ** we are manually writing down SKU codes on each waybill**. The reason why we do this is because not all people in the production floor knows how to read full English SKU descriptions with 100% comprehension, since English is not our native language. So instead, we had to come up with SKU codes so as to write them into individual waybills so anyone can understand,

So I spun up my very first major contribution to our business: Waybill Reader and Writer.

This is where the PDF uploads occur

And this was what it is doing. *Read some more, I promise you, this gets more interesting. ;)*

Intially, I created this suite to process pdf waybill files so as to eliminate the time spent writing in every single page of waybills (which can span up to 300 pages per file)

And then, ** that was supposed to be it**. We were happy, the tiny CTO in me got excited that I managed to integrate, even a tiny suite, to save 50% time in manually writing stuff.

Then, I realized: *"If I was already processing and reading orders in each waybill, that means I can read data for each and every single order that comes in, right?"*

Then, another ** eureka** moment came in.

Since we are an e-commerce business, **managing inventory is a crucial process**, so as to avoid discrepancies and suprises later on.

Initially, I integrated a daily inventory count for each SKU.

Then, we would also want some granular level inventory as well in order to use these information to further help us scale, identify top-sellers, etc.

This again, adds up to the total time saved! Instead of doing an actual count every week, we can get an updated inventory daily, and match it with the actual count that we decided to do on a monthly basis instead.

**Total time saved: 75%**

Coming from an era where I started with ASP.NET, Jquery, AJAX, Web services, all the way to the era where front and back-end stuff are separate, DevOps is a separate role too, which was a surprise for the 2019 year-old me. Github Copilot was a breath of fresh air!

Being able to ask anything, make Copilot do something, how to do things efficiently, and step-by-step deployment **(I am sure a lot of devs had a hard time with this part)**. Bottomline was, Github Copilot was very thorough in writing documentations to make sure I don't get lost deploying stuff.

Having the * knowledge of tech jargons* (ssh, droplets, logs, screen, etc), it helps a lot so as to save time having Github Copilot understand what you actually wanted to do.

I can safely say this: * If you know your way around things as a dev, Github Copilot is a great assistant* to be able to optimize your workflow, be it your software architecture, CI-CD, documentation, you name it.