Uncategorized
Segment the audience and focus on personalizing services for the most profitable group of clients
About the client and the pursued goals
966.ua is one of the largest Asian food delivery chains in Ukraine. Supported by the European Bank for Reconstruction and Development, they turned to us for a solution they could use to monitor their business metrics.
The solution
We designed a custom dashboard with four main data sets: Finance, Clients, Production, and Web Analytics.
By choosing tabs and filters, it became possible to gather data from Google Analytics, an internal accounting system, and external sources in one place.
All the executives can access the data they need – such as income and expenditure dynamics by cities and departments, marginality and net profit, monthly revenue per unit, product rates, repeat purchase rate, lifetime value, churn rate, etc. – whenever they need it.
The dashboard made it easy for them to segment the audience and focus on personalizing services for the most profitable group of clients.
Product set analysis
Thanks to the formation of associative rules, we analyzed two years’ worth of data on transactions and discovered what product combinations are the most popular among 966.ua’s clients. Special graphs were designed to present visually which products are bought in sets with other products and patterns in users behavior. This report helped the company to improve its product set promotions.
Product set analysis
Thanks to the formation of associative rules, we analyzed two years’ worth of data on transactions and discovered what product combinations are the most popular among 966.ua’s clients. Special graphs were designed to present visually which products are bought in sets with other products and patterns in users behavior. This report helped the company to improve its product set promotions.
Client segmentation
This report was prepared for the food delivery company to identify client clusters for better personalization and to allow it to focus on the most profitable group of users.
On the graphs, we presented the frequency of orders and the average purchase price of every cluster, the income from the customer, and time of the order.
Every cluster was studied separately and its share of the whole customer base identified.
