A look at some of my top success stories as a Data-Driven Marketer

Growth Marketing Analysis for ASAP Delivery

Objective

To conduct a thorough analysis of user retention and activation for the ASAP Delivery App to gain actionable insights aimed at boosting engagement and interaction within the app.

Problem

ASAP's Product team identified that, for two consecutive years, the volume of active users remained stagnant, despite ongoing Paid Media acquisition efforts bringing in new users each month.

Solution

Leveraging their multiple data sources and internal data lake, I ran an ETL process to obtain clean, relational tables in SQL. Then, I performed an Exploratory Data Analysis to understand both the data and the business logic. I created Growth Marketing charts, such as the Retention Life Cycle, to pinpoint users who maintained a steady buying volume. Additionally, I set up Retention Cohorts to highlight events impacting various user segments' behaviors and helped define metrics that projected whether a user had reached the Habit Moment, Aha Moment, and Setup Moment.

Results

Following this analysis, I identified specific points in the Customer Journey where users dropped off and uncovered the key reasons users failed to develop a purchasing habit in the app. These insights were communicated to the ASAP team, who implemented the recommended changes. Six months post-implementation, a follow-up analysis confirmed a total retention improvement of 18% from the initial retention rate.

*Data and insights in the following charts have been randomized to maintain client confidentiality

"Great analysis! Thanks to these findings, we can better understand our users and implement growth strategies that drive more engagement with them."

— RODERICK A., ASAP DELIVERY CEO

Predictive Lead Scoring with Sci-Kit Learn

Objective

To develop a tech solution to improve the efficiency of tracking a large base of potential clients for the sales team, without increasing operational costs.

Problem

The high volume of leads received monthly made it difficult for EDUCET’s sales team to properly follow up with users interested in purchasing an online course. Additionally, EDUCET couldn't afford to expand its sales team or reduce acquisition efforts.

Solution

To address this challenge, I used the logistic regression algorithm from Sci-Kit Learn in Python to create a predictive model that assigned a score from 0 to 100, indicating the likelihood of each lead converting to a sale. This model was powered by metrics gathered from user behavior on the company’s website, like time spent on the page, visited URLs, and information submitted through forms.

Results

The final version of the logistic regression model achieved an accuracy rate of over 82% when tested in real conditions.

Six months after implementing the predictive model, the marketing team confirmed an increase in the lead-to-customer conversion rate from 5% to 14% per month, as the sales team could now focus their efforts on leads with a higher probability of conversion.

*Data and insights in the following charts have been randomized to maintain client confidentiality