Overview

ABC Company (“The Company”) is a new delivery service, predominantly operating in Northern California. The Company primarily operates in ethnic food category, specifically Chinese food.

Objectives

The Company has ran into some growth challenges. They are looking to capture more market share in the near future but had trouble determining the best strategy for growth. The Company reached out for help to understand the bottlenecks and roadblocks for growth, as well as determining an actional stratgey to achieve the overarching goal.

Data Provided

The Company provided a set of delivery data for the purpose of this analysis. The dataset includes all information pertaining to a specific order.

Step 1: Data Cleaning

The dataset provided contain large amount of information relevant and unrelevant for the purpose of my analysis. The first thing I wanted to do was to make sure that there was no duplicate, as well as getting rid of data outside of the observation period. I also had to transform some date-format information into integers, which will later be beneficial for mathematical computations.

Step 2: Setting Assumptions

In order for your strategic recommendations to work, you need underlying conditions to support them. These are the assumptions you base your hypotheses on. If these assumptions don’t hold true, your recommendations will fail. That’s why clearly stating assumptions is critical—you’re constructing a hypothetical world around them.

Some of the key assumptions I anchored my recommendations on for this case:

• The shorter the total wait time, the higher the user satisfaction. Higher satisfaction leads to lower churn and higher referral rates.

• Customer tip amount is positively correlated with driver efficiency. Higher tips as a percentage of order total indicate higher efficiency and user satisfaction.

• User satisfaction is positively related to the number of restaurant options available.

• The dataset contains an exhaustive list of restaurants available on the platform.

• The number of orders a restaurant receives is positively correlated with its ability to deliver user satisfaction, popularity, and quality.

• The amount a user spends on the platform is positively related to their loyalty.

• The more a user spends, the less likely they are to switch to a competitor.

• Order frequency patterns are the same for The Company and its competitors. If 6 PM is the most popular time, it is also the peak time for competitors.

• The amount a customer requests in refunds is negatively correlated with their satisfaction with the platform. Higher refunds indicate lower satisfaction.

Step 3: Data Analysis

I performed the analysis for this project strictly on Excel. Certainly, since the dataset was provided in csv. format, I could’ve used other data analysis tools such as SQL and Python. But honestly, hey, don’t judge me, but I am sucker for Excel coming from a background in Finance. Whatever gets you there, am I right?

For my analysis, I relied heavily on INDEX MATCH, VLOOKUP and similar functions to consolidate information acorss different criteria. Through analyzing the dataset, I wanted to gain a better understanding of the current delivery speed, top 10 drivers / customers / restaurants on the platform, and the most popular ordering time. From there, I can orient my strategic recommendations around them.

Example: Analyzed outcome of most popular delivery time

…A DEVELOPING STORY…