Client context
Thames Water is Britain’s biggest water supplier, custodian of a network of sewers and mains more than 200 times the combined length of all the UK’s motorways. In a regulated industry where customer satisfaction directly affects public funding, the company needed to make better use of its data.
The business problem
Thames Water wanted to improve its customer satisfaction scores, which are benchmarked by Ofwat against other operators. Raw data suggested rapid fluctuations, but it was difficult to distinguish genuine trends from statistical noise. The company also needed to predict equipment failures across a geographically diverse asset portfolio, and reduce the risk of sewer flooding.
What we did
Red Olive performed key value driver analysis to identify which tasks were dragging down customer scores, enabling Thames Water to allocate resources where they would have the greatest impact. We profiled each network site individually, grading pumps and pipes to predict likely failures based on historical data and local conditions.
For flooding risk, we built predictive models combining static data (topography, pipe geometry) with dynamic data (weather, maintenance schedules) to produce heat maps highlighting locations where flood risk was increasing.
The results
Thames Water gained a clearer picture of customer satisfaction drivers and could allocate resources more intelligently. The predictive models enabled proactive asset management and the flooding risk models allowed maintenance teams to be dispatched before problems reached danger levels.
Looking ahead
The models and analytical tools continue to help Thames Water forecast issues before they arise, protect customer satisfaction scores and secure better returns on infrastructure investment.
