Analysis of User and Merchant Dropoff for Sugar App Time Series

Home

Project Overview

Findings

Project Documentation

Project Management

Mid-Term		Finals

Funnel Plot Analysis		Time Series Analysis		Geospatial Analysis

Abstract

Our client is a city guide discovery application that brings users and merchants together through geo-located offers. The objective of this research is to examine merchant performance via redemption rates. The results are displayed using funnel plots, a useful tool for displaying unbiased information on performance outcomes when comparing entities within a group. The funnel plot shows a high amount of overdispersion where there is a large number of outlying merchants. By further analyzing under-performing and over-performing merchants separately, the analysis shows that there is also a large variation in outlying redemption rates within each group. To investigate the underlying reasons, we conducted exploratory data analysis. Merchant and product category are shown to be significant contributors to a merchant’s redemption rate. These findings will help our client set benchmarks for individual merchants and develop interventions to help merchants increase their performance.

Business Motivations and Objectives

Methodology

Data

Data Preparation

Tools Used

Constructing the Population Regression Model

Method for Hypothesis 1

Method for Hypothesis 2

Method for Hypothesis 3

Results

Hypothesis 1: Merchant Growth(IV) is associated with User growth(DV)

Hypothesis 2: User Growth(IV) is associated with Merchant growth(DV)

Hypothesis 3: Revenue Growth is a function of User and Merchant Growth

Discussion

Implications

Prediction Model

Univariate Prediction Model

Multivariate Prediction Model