Case study

Customer & Product Insights and Optimisation

Fashion, Retail

Forecast had a critical impact on our client's operations by advising on the move from local data storage to scalable cloud storage, and by unlocking the commercial value of this data through advanced analytics.


Our client is a leading global online fashion retailer group comprising of 13+ brands and an annual revenue of $1bn+, spanning multiple markets across Europe, North America, and Australasia regions with a 14m+ active customer base. In the fast-paced fashion industry catering to the ever-changing audience taste, our client utilizes today’s technology to maintain efficient operations across customer, product, and pricing sectors.


The client required support to achieve commercial goals across multiple sectors with a data driven strategy. Despite the abundance of raw transactional, product, and financial data accumulated in the last few years, our client was still early in their data journey, starting to develop a cloud-based scalable data infrastructure to generate value through analytics. Forecast worked with the internal data analytics team to add value through insightful analytics as well as advanced machine learning solutions across three business lines.

Key Deliverables and Outcome

Following is the non-exhaustive list of where Forecast was able to add material value to our client’s operations.

Customer Centric:

  • Customer segmentation and marketing analysis per brand to focus on better understanding of customer groups, leading to targeted marketing that drives customer retention.
  • Calculated Customers Lifetime Value based on spend history and aggregated churn probabilities, driving company focus on increasing high-value customer counts.
  • Identify and understand impact of premier product offering, along with optimal pricing and target audience for future acquisitions.

Product Centric:

  • Built dashboards that feed from data pipelines and report on primary KPIs across the product range, delivering daily insights to stakeholders.
  • Developed an end-to-end item level machine learning predictive solution to model likelihood of products getting returned by customers. Model predictions feed into accurate forecasting reports as well as drive merchandiser’s decisions on restocking product in warehouse.
  • Understanding the P&L performance of products at peak and regular sale periods. Delivered insights for stakeholders to make decisions on pricing and promotion of products.

Along with the above, Forecast also provided the following across multiple client teams:

  • Assisted internal data team to move from legacy local data storage to scalable cloud-based solution.
  • Built and maintained multiple SQL and Python ETL pipelines across Qlik and Google Cloud Platform, enabling internal teams to process daily reports with better efficiency.
  • Worked with multiple teams to process 1bn+ rows of data across dozens of datasets, introducing good analytics standards to the internal analysts.

More Case Studies

Case study


Budgeting & Acquisition Model

Case study

Real Estate Finance

Loan Reporting Model​

Loan Reporting Model​

More Case Studies Contact us