AI FORECASTING REPORT: What Are the Challenges to Predictive Modeling


By Terry Kades, BlueWave Forecasting

It’s a fair assumption to say that AI methods and their algorithms are pretty well nailed down and optimized.

The engineers have done well. But that’s only half the AI power shake recipe. The power of AI comes from the ingredients in the food it consumes. It sounds a little crazy, but read on for just a minute and see why the AI diet matters so much.

Some AI systems e.g. ChatGPT require words, language rules and dollops of subject matter (food) to guide it. The areas such as forecasting revenues, product volumes, marketing KPIs etc. it’s not words but values that AI systems must consume, crunch and deliver real world value that people can use every day.

AI needs both historic data AND data of the drivers that impact different scenarios and forecasts.  We are talking about causal drivers and not just things that are correlated.

Data is AI food. In people’s food, there is junk food and nutritious food. Some junk is made to look and taste good but it’s unhealthy. The same is true for data feeds into AI tools because only

‘Nutritious” data can provide useful results. Junk data will infest the system and require processing to identify the junk and send it to the wastebasket. So rather just do it right and feed nutritious / relevant and meaningful data in the first place.

In the context of AI for forecasting, what does nutritious data look like, and where does it come from? Historical data is part of the answer. It’s like water, our bodies require it to exist, but we also need a variety of food to exist, to stay healthy and to be productive.

What is “variety of food” and “nutritious data” in the context of AI forecasting?

Simply put, it is the drivers and their data. Things that drive the market, the industry, the business. But make this information usable by AI tools. And therein lies the challenge and the reason why AI forecasting is lagging.

To be usable in AI, data feeds need structure and perspective and there is nothing better than specially formed scenarios that are built to cater to AI’s requirements.

Terry Kades is the CEO and co-founder of BlueWave Forecasting, which provides industry and business scenarios and forecasting. An experienced director and senior management consultant he is expert at identifying the drivers and creating scenarios that guide business plans, enhance analysis of investments and provide accurate forecasts. He is convinced that the true power of AI will flourish when it incorporates properly constructed scenarios that fertilize and boost the intelligence aspect of AI.