ANLY482 AY2017-18 T1 Group1: Project Overview
Our project sponsor, eatigo, currently has 2 million registered users. However, reservation and reactivation rates remain low as only 30% of eatigo's subscribers have ever made a reservation. Further, even amongst this 30%, each customer makes only 1.6 bookings on average per month. Our sponsor believes that there is higher potential, given their presence in markets such as Singapore and Thailand, both of which are amongst the top 3 spenders in South-East Asia when it comes to dining out.
Eatigo’s business model is such that it earns revenue when customers show up for their booking. For every customer that does show up, Eatigo receives a certain fee from the vendors. Therefore, it is crucial for eatigo to have a good vendor listing (so customers are interested in making their booking through eatigo) and ensure that customers are aware of these vendor listings and discounts available (so customers are incentivized to make their booking through eatigo).
Through our project, aim to improve reservation rates amongst subscribers who have never made a booking and reactivation rates amongst subscribers who have made a fewer number of bookings per month. We plan to do this by analysing transaction, redemption and vendor data to reveal booking patterns across customer & vendor types.
As described, the overall objective of our project is to help eatigo increase number of bookings per customer. With this in mind, at this stage, our aim is to conduct an Exploratory Data Analysis (EDA) to understand the vendor choice, booking and redemption patterns of eatigo customers across its 8 regions, and 700 restaurants.
From our EDA, we would like to answer the following questions from our datasets :
1. Analysis of Vendor Data
- Variation of booking patterns across cuisines
- Variation of booking patterns across neighbourhoods
- Variation of peak and downtime across restaurants
- Understanding the potential of optimizing discounts provided by restaurants within the same neighbourhood
2. Analysis of Customer (Transaction & Redemption) Data:
- Variations of booking patterns across customers that have made bookings previously
- The average gap between the time booking is made and the time customers have to show up at the restaurant
- The customer clusters by restaurant type, location, timing and discount preferences
The current objectives may be subject to further changes after we have obtained the actual data.
We are waiting on the data from the sponsor, which we will be receiving once the Non-Disclosure Agreement is signed.
We have currently requested for the following categories of data:
- Vendor Data
- Transaction Data
- Redemption Data
Some of the variables that we expect to receive in these datasets include the following:
| Vendor Data | Transaction Data | Redemption Data |
|---|---|---|
| Restaurant ID | User ID | User ID |
| Country | Restaurant Booked ID | Restaurant Booking ID |
| Restaurant Name | Country | Country |
| Cuisine | Number of Diners | Number of Diners |
| Location (Neighbourhood) | Date & Time of Present Booking | Date & Time of Present Booking |
| Opening Hours | Date & Time of First Booking | Discount Received |
| Peak Hour Time | Booking Type | Promotion Code |
| Planned Discounts | Booking Status | Booking Status |
| Maximum Capacity | Booking History | Showing up History |
| Cuisine Preferences | Cuisine Preferences |
References:
“Singapore Among Top Spenders, Asia Pacific Survey” (http://www.todayonline.com/singapore/singapore-among-top-spenders-asia-pacific-dining-survey) “How Eatigo has Disrupted the Food Space.” (https://ecommerceiq.asia/eatigo-food-commerce/)
