Data Insight works with clients to transform the way their businesses operate through the use of advanced data science techniques including machine learning and artificial intelligence.
Using our proprietary AI driven tools and algorithms, DI Predict can help you better understand your customers including predicting churn, upsell opportunities and new product uptake.
DI predict helps our customers in the following areas:
TIME SERIES PRICE/DEMAND MODELLING
Understanding how price or demand is likely to change over time, based on what has happened in the past.
A client was launching a product in a geographical area where it had not been previously available. Based on uptake in similar areas, Data Insight forecasted demand and provided dashboards illustrating how well uptake matched our predictions over time.
NATURAL LANGUAGE PROCESSING
Mining text data (such as survey responses, NPS data and chat-bot entries) to understand common themes and sentiments.
Our client had built a digital assistant, and needed to know what their customers were asking her. We used topic modelling to cluster past conversations and find the major topics. From this, we found opportunities to get a human in the loop faster, if one was needed to solve the problem.
PROPENSITY MODELLING
Helping clients to predict the likelihood of a certain event occurring (such as customers churning or taking up a product) and using that information to inform targeting marketing.
Data Insight built a churn propensity model for a large fuel retailer to address retention in their B2B customer base. The model helped the organisation to identify customers most at risk of churning and optimise marketing spend to focus on the biggest priorities.
CAMPAIGN EVALUATION AND CHANNEL OPTIMISATION
Understanding business impact of marketing campaigns and which channels are most effective at stimulating activity.
Data Insight provided statistical evidence to a client regarding an increase in the uptake of a particular product due to their marketing activity, measuring the effectiveness of the campaign.
Di PREDICT proprietary churn models enable clients to understand their customers reasons for churning and take action before its too late. They are customised across key industry verticals.
.png)
