Sales Forecast using Facebook's Prophet
Rafael Duarte
Overview 🔎
Forecasting demand is a must for all businesses. When done properly, it can help save and optimize resources and time.
With that in mind, I set out to create this project based on and actual demand received by our company. To preserve the client's identity and data, I used publicly available data to come up with this case study for my Portfolio.
Problem & Solution 🤝
In a nutshell, the goal of the project was to predict sales for a given period. With that information, the business would be able to:
• Improve stock and, cashflow management. • Price items correctly, to make sure the right products are sold at the right price at the right time. • Better understand the seasonality of its products.
Process 🛣
All of that was achieved in a simple manner, taking advantage of Facebook's Prophet library, which provides a simple-to-use solution for Time Series problems, such as this one.
With just a few tweaks and lines of code, I was able to create a robust solution that provided the business with useful insight.
Results 🎁
With the results from the project, the client was able to have reliable insights, which helped them optimize their operation, which in turn resulted in more profits.
Takeaways 📣
This project shows that it doesn't take overly complicated algorithms to make an impact in a business. Furthermore, it also showed the importance to have a deep understanding of your own data.
