This repository contains project documentation and output of market insights and recommendations for local snack (seblak) sales using web scraping techniques on Tokopedia.The project utilize documentation in Bahasa Indonesia and may presents specific or local/industrial terms related to the e-commerce industry.

🚩 Project Name: Predicting Airline Customer Satisfaction
🙋🏻‍♂️ Project Owner: Nicku R. Perdana
🏁 Date Finished: Feb 08, 2023
📞 Contact: LinkedIn; E-mail

Situation:

When income is still insufficient to meet daily needs, starting a business becomes one of the viable options. I am considering opening a dropshipping business with the following considerations:

• Selling without the need for a personal warehouse: The buying and selling process can be done remotely without the need to manage inventory further.
• Freedom and flexibility: Being able to choose products to sell based on current best-selling trends to enhance profitability.
• Business without sacrificing time: Dropshipping business can be a solution to supplement income without sacrificing primary employment and starting from scratch.

Complication:

Dropshipping business can be tricky. Data analysis skills are tested to determine trending and highly profitable merchandise with various influential indicators. Based on initial identification, the instant noodle snack 'seblak' is currently popular among various age groups. Besides being a popular consumption product, the consumptive nature of Indonesian society - consumption activity as the main driver of the country's economy - also makes this commodity have good opportunities and stable trends.

Question:

Several questions are sought to be answered in this analysis process as follows:

• What is the minimum and maximum income potential that can be obtained from dropshipping 'seblak'?
• From which locations do stores tend to generate profitable income?
• Is it necessary to buy products for dropshipping from stores with high ratings?

Answer:

To answer the following questions, the steps I can take are to set problem-solving as a goal using the S.M.A.R.T. framework:

• Specific: Increase monthly income by running a dropshipping business on the Tokopedia e-commerce platform.
• Measurable: Increase monthly income by at least Rp5 million each month.
• Achievable: As a Data Science Hacktiv8 boot camp graduate, the sequential steps to take are as follows:
  • Analyze the current situation of 'seblak' product sales on Tokopedia by using web scraping techniques on the first 10 pages of 'seblak' product searches on Tokopedia. This technique allows me to gather data related to the list of product names sold along with prices, quantity sold, store names, store locations, and buyer ratings.
  • Clean the data resulting from web scraping into a table format that can be explored, processed, and analyzed.
  • Conduct descriptive statistical analysis to produce conclusions about the distribution and trend of price data, quantity sold, and buyer ratings of 'seblak' products.
  • Perform inferential statistical analysis to draw conclusions about which products are sold the most, potential best-selling products, and profit potential.
• Relevant:
  • 'Seblak' is one of the snacks/products that is currently trending and being discussed in various online and offline domains.
  • Online shopping becomes a solution to obtain items that are only available/more affordable to obtain from certain areas.
• Time-bound:
  • This goal will be set to last for the first 3 months since starting the dropshipping business to provide time to achieve the additional monthly income target of Rp5 million.

Findings and Conclusion:

Based on a series of observations and data analysis conducted, the display data of 'seblak' products sold on the Tokopedia platform tends to vary. Some are sold from Rp350 to as high as more than Rp200 thousand. This indicates a non-normal distribution price trend, leaning towards the left. It means that only a few items have extremely high or low prices, which are outside the general market price pattern. Meanwhile, the kurtosis test findings also indicate that most products are centered on a specific price point, proving that the 'seblak' price has a single market equilibrium and can have special extreme price cases, but with few products.

This may occur because 'seblak' products, besides being generally sold in packaging, also indicate wholesale/bulk sales. It should be noted that by querying the word 'seblak' on the Tokopedia platform, there are many possible search results that will appear, such as those who only sell 'seblak' raw materials, seasonings only, and even the packaging. These can potentially be included in the data collection results using web scraping techniques.

By using the assumption of monthly income from selling 'seblak' by multiplying prices and quantities sold, the potential income that can be earned varies. Therefore, cleaning outlier values ​​from the multiplication of these two variables gives the conclusion that at least the monthly sales income ranges from Rp6,153,331.62 to Rp6,157,210,79. This aligns with the main goal of wanting to get additional income by selling 'seblak' by at least Rp5,000,000 each month for the next 3 months.

In execution, dropshipping will not consider store location as the basis for determining which products to purchase. The analysis found that the price of 'seblak' products is not affected by the seller's location. This is crucial so that issues regarding timing accuracy and distribution will not affect the business process in the future.

The viral nature of 'seblak' products will also be considered in taking the main business steps, by looking at the popularity aspect of a product. The findings show that buyers tend to prefer cheaper 'seblak' products. Spearman's Rho correlation test indicates that the cheaper the 'seblak' product, the more likely it is to go viral and be popular. This will greatly support the planned business prospects. In the future, the 'seblak' business will feature cheaper products that have been highly rated by previous buyers.

---