Severn Trent transforms energy forecasting and predictive planning with CCH Tagetik Expert AI
Key Takeaways
- 90%+ forecast accuracy
- 36× faster model training
- 24M+ rows of data analysed
The challenge: Improving forecast accuracy for a £200M energy spend
As one of the UK's leading water utilities, Severn Trent manages approximately £200 million in annual energy spend while operating in a highly regulated industry where cost efficiency and service quality are business-critical.
Accurately forecasting energy costs is inherently complex. Consumption is influenced by dozens of operational, financial, environmental, and market variables that constantly change over time.
Historically, these data sources were spread across multiple systems, making it difficult to bring them together into a single forecasting model. As a result, finance teams lacked full visibility into the business drivers behind energy consumption and future costs, limiting both forecast accuracy and decision-making.
The solution: Bringing finance, operations and AI together for Predictive Planning
Building on its existing CCH Tagetik platform, Severn Trent deployed CCH Tagetik Expert AI to create a unified and explainable forecasting model.
|
The project consolidated:
into a single planning environment. |
The model evaluates multiple business drivers simultaneously, helping finance teams understand not only what is likely to happen, but also why it is happening. This enables a more intelligent and data-driven approach to predictive planning, where finance can continuously adapt forecasts based on changing business conditions.
Why it matters: Moving beyond traditional forecasting
Traditional forecasting often relies on historical financial data alone.
Predictive planning becomes significantly more valuable when operational and financial data are analysed together. Rather than relying solely on historical trends, finance teams can continuously adapt forecasts based on the business drivers that matter most.
The business impact: Higher forecast accuracy, faster decisions and better business outcomes
90%+ forecast accuracy
By combining financial, operational, and environmental data, Severn Trent is targeting 90%+ forecast accuracy, improving confidence in planning and budgeting.
36x faster model training
Model training and execution times were reduced from 12 hours to just 20 minutes, enabling faster forecast updates and more agile decision-making.
Hidden drivers brought to light
Expert AI uncovered previously invisible relationships within millions of rows of data, helping teams better understand what drives energy consumption and costs.
Faster, more informed decisions
With quicker access to insights, Finance can focus less on reporting and more on supporting business decisions with confidence.
-
What are Severn Trent now be able to do differently with CCH Tagetik Expert AI?
What we're able to do now is look at how, even as a finance function, we take advantage or we utilize the operational data, that the business consumes and how to then finance and learn about that data and then use it in a seamless way to deliver what would ultimately lead to better financial forecasting moving forward. So they've got quicker answers to their questions. Their decision making can be done quicker. They'll learn more about what's going on within the business. -
Why did Severn Trent choose CCH Tagetik for predictive planning and AI-powered forecasting?
I've really valued the level of support and input we've had from CCH Tagetik. I would kind of certainly advocate other people for, looking at the predictive intelligence tool. The machine learning AI embedded analytics that's part of the suite because we're really confident that it's only going to improve; it's going to get better. Being a single platform, we've already got Consolidation, we've already got FP&A, and now we've brought the AI piece into that. The Analytical Information Hub has been really useful for that.