With o9 Demand Planning, planners can improve accuracy and reduce the time spent in forecasting. So how do we make this happen? It's through a combination of workflow, data and technology. All these three elements can be brought together in the o9 Digital Brain.

The digital brain converts data into knowledge by connecting large amounts of discrete data using "nodes" and "edges" to build the Enterprise Knowledge Graph, or EKG. o9 machine learning uses this knowledge, based on lagging and leading indicators to make better predictions. o9 also highlights actionable exceptions for planners to make faster and more insightful decisions. Let's look at some examples of o9's intelligent predictions, and insight based on internal and external driver data.

Here we can see from ILI activity, which is influenza surveillance data, that there is a strong correlation with the sales and forecast of cold and flu relief products. This graph shows how Sell Out data has a fairly strong correlation with search trends from Google. Consumers are clearly looking for the product in question. Here we see that Sell Out data very strongly correlates with glance view data.

Glance views are the number of distinct views or clicks that a product receives on a website. It's clear that you could place a high level of trust in glance views. Next, let's look at the impact of list prices. Here we can see that the sales dropped significantly for this product in the month of September.

Bringing in the list price into the picture shows that the drop in sales was tied very closely to the price hike in the same month. The o9 Digital Brain learns from this feature and uses this relationship to predict a drop in volume for the next price increase planned in 2021. Including historical prices and expected price changes can improve forecast prediction and minimize excess build. Now let's consider the impact of short shipments or cut quantities.

This example shows shipment history with a crazy volatility. If we load the cut quantity data, you can clearly see the missed shipments that were due to supply challenges. Simply adding the cut quantity to shipments will over inflate the demand because customers re-order the same quantities if the original order was not met. o9 Digital Brain can correct the history by using a more sophisticated algorithm, such as a seasonally adjusted linear interpolation to reflect the true demand.

This corrected shipment history can then be used as an input for forecasting. As you can see from these examples, the o9 Digital Brain with machine learning prediction can dramatically improve forecast accuracy, and provide actionable insights and significantly reduce the time that Demand Planners spend in the forecast consensus process.