Aside

Diego Usai

Contact Info

Skills

Main

Diego Usai

Customer Insight | Business Science | Marketing Analytics

Experienced Analytics Professional, Consultant and Data Manager with expertise in Project Management, Team Leadership and Stakeholder Management, and an insatiable curiosity about how to use data to inform business decisions

Professional Experience

Self-employed

Data Scientist

London, UK

Jul-18 – Present

  • Working on Customer Analytics & Business Intelligence projects
  • Publish technical articles on Towards Data Science
  • Complete Business Science University courses

Medium - Towards Data Science

Technical Writer

Online

Mar-19 – Present

  • Publishing articles covering topics in Machine Learning, Business Intelligence, Customer Behaviour and Insight, Data Collection & Wrangling, Segmentation & Cluster Analysis, and more to come!

Ebiquity plc

Digital Data Manager

London, UK

Feb-17 – Jun-18

  • Lead and mentor a team of 3 Data Analysts
  • Collect and maintain data from various sources (Ad Server, Third-party Verification, Google Analytics, Shareablee Analytics, etc) to analyse and present to clients to automotive, banking and e-commerce clients
  • Develop and maintain R Shiny dashboards to visualise large amount of sales and online marketing data and conduct regular market trend analysis
  • Use Structural Equation Modelling in R for Social Sentiment Analysis to assess impact of Paid, Owned and Earned media mix on social consumer sentiment for Subway restaurants chain

Ebiquity plc

Senior Effectiveness Consultant

London, UK

Apr-13 – Jan-17

  • Lead consultant across national and international clients, working collaboratively with stakeholders to understand issues, translate them into insight briefs, and use results to shape strategy
  • Work with RIAS Insurance stakeholders across several business units to identify business relevant customer segments, carry out time series analysis to understand key drivers of customer acquisition and retention for each segment, and use learnings to inform future strategies
  • Use SQL and R to collect and compile structured and unstructured data from various sources to run bespoke analysis and generate insight
  • Run media budget portfolio optimisation and what-if scenarios to produce sales forecast based on optimum marketing budget allocation
  • Work with stakeholders to design, execute and analyse results of several online and offline media testing to answer clients’ specific questions

Neustar MarketShare

Marketing Scientist

London, UK

Feb-13 – Mar-13

Annalect

Consultant Analyst

London, UK

Feb-12 – Jan-13

  • Lead consultant responsible for scoping out, setting up and delivering a brand new marketing mix model offering for Sainsbury’s retail
  • Contributing author to the Silver Award Winning IPA 2012 Effectiveness Paper “Waitrose: The best ever Christmas“

Havas Media Group

Effectiveness Director

London, UK

Sep-10 – Jan-12

Publicis Groupe

Senior Econometric Analyst

London, UK

Aug-08 – Aug-10

Gain Theory

Analyst

London, UK

Apr-06 – Jul-08

Education

Business Science University

Business Analysis with R (101-R)

Online

2018–2019

  • A data science for business course focused on clean, join, manipulate and visualise data with the Tidyverse suite and the Recipes package, modelling with Parsnip and reporting with RMarkdown

Business Science University

Data Science for Business (201-R)

Online

2018–2019

  • A data science for business course focused on solving a real-world churn / binary classification problem
  • Modelling & performance analysis with H2O AutoML and LIME, link Machine Learning to Expected Financial Performance and develop a recommendation algorithm from scratch

Università degli Studi di Cagliari

Laurea in Economics & Statistics, 1:1 Honours Degree Equiv.

Italy

1994–2000

  • Part Time, specialised in Time Series Econometrics & Forecasting
  • Dissertation - “Non-linear Econometric Models: an Application to Modelling and Forecasting GDP”

Selected Projects

Modelling with Tidymodels and Parsnip - A Tidy Approach to a Classification Problem

I this project I use R tidymodels to create and execute a “tidy” modelling workflow to tackle a classification problem

N/A

2019