LLamasoft Supply Chain Blog

← Back to Blog

Podcast: Evolve Your Demand Forecasting Process

By Matt Gunn  September 16, 2019

With constantly changing market conditions, heightened customer buying expectations and shorter product life cycles, it is almost impossible to predict demand without incorporating external dynamics like these into the demand forecasting process.

Listen to the podcast: 

In LLamasoft’s first podcast, Senior Director of Product Management, Vikram Srinivasan, talks with me about the importance of taking an “outside-in” approach to demand modeling to help companies improve their forecasts and understand key factors that influence demand. We discuss how demand, data and advanced AI capabilities can reveal new possibilities through interconnected insights.

Approaching eight years in his tenure with LLamasoft, Srinivasan has a background in operations research and a passion for innovating and bringing new products to market. He has a deep perspective on demand modeling, forecasting and the challenges that businesses face to predict the future.

One of those challenges is though companies have been forecasting and modeling demand for some time, they often base it off of one variable – demand history. But there are multiple factors to consider to be successful. “There are about 10 to 15 more variables that are going to impact your demand. If you don’t consider those factors, then the forecast only solves for 1/10 of the problem,” Srinivasan says.

Some of the variables he’s speaking about are industry-specific. “In retail, it’s all about point of sale, promotions, price and consumer price index. In food and beverage, temperature can be a highly important variable.” But he explains that most factors are common across many industries such as macroeconomic indicators like GDP, consumer price index and consumer confidence. Companies can begin to consider those factors to understand what drives many supply chain, product and marketing, and longer-term strategic investment decisions.

Another blunder companies encounter is becoming comfortable in predicting a pattern of growth in sales and increased volume. These combined misinformed practices can cause inaccuracies in planning which then causes inefficiencies in supply chain decisions in the long run. “Change is the only constant. If you can’t account for changing business and external factors in your forecasts, the implications can be detrimental to your business. There are a lot of investments you need to make in your supply chain to stay ahead of the market and the competition,” Srinivasan suggests.

Companies need to understand gaps and use an outside-in approach to provide value to customers. Srinivasan says, “As soon as you open yourself up to understanding where the market’s at and the importance of an outside-in approach to forecasting, I think the rest of it is pretty quick, honestly.”

Upon gaining insight, use AI and human expertise to quantify external factors. “With AI, you gain a good understanding of key variables that are highly correlated to your demand right out the gate. When you also bring in the right domain experts, they can look at this list and tell you which ones make sense to include and which ones don’t,” Srinivasan advises.

Within two weeks of setting up LLamasoft’s Demand Guru from start to finish, Srinivasan states, “our customers quantify the improvements and see the exact implication to their supply chains.” He has witnessed companies leverage AI and external data to reach a 15 to 20% forecasting accuracy improvement. The key is this – it’s not just about improving the demand forecast, but also monitoring causal factors on an ongoing basis. He points out, “As those factors change, demand planners also understand that their future’s going to change. If you get ahead of that, then your forecasting will also be proactively managed for the mid- and long-term.”

In essence, Srinivasan concludes, “Outside-in forecasting, coupled with the right amount of applied AI and machine learning, I believe can really change the game for how you make those long-term decisions with more strategic precision and certainty.”

Learn more about LLamasoft’s demand modeling solutions and how you can better predict the future for savings, efficiency and growth.