Is Supply Chain a chain? It would be more appropriate to be called a web or a storm of nodes with supply and demand as the major points. Before we go into how Data Science in CPG Industry is helping deal with a few of the supply-demand issues, let’s talk a little about the current situation and what is happening in the CPG Industry.
Let’s explore the demand side first – In addition to the pent-up demand after pandemic-induced lockdowns, increase in the fast-delivery apps, and rising inflation, the fluctuations that the CPG industry faced is parallel to no other.
In order to understand the supply side, let’s take a look at the holiday season of 2021 went. According to a study, 98% of surveyed retail executives said they were experiencing supply chain issues before the holiday season, and they were expecting it to last at least till 2022. Even now, some ports are experiencing more than 20% congestion due to lockdowns.
All of this is affecting the products in a way that the cost is increasing, which is making executives increase product prices or shipping costs. Other than these what can executives do? Reduce the number of SKUs, expand the supplier base, and want more visibility into the inventory levels. Let’s talk more about how Data Science can help CPG companies at least in the short run to deal with such ongoing challenges.
Helping CPG Industries with Data Analytics and Data Science
-
Understanding Product Portfolio
Identifying what are the products that are working out or what aren’t is the first step in leveraging the analytics. The multiple SKUs in a product can be discontinued or the complexity can be reduced based on personalized scoring, based on
- Supplier Risk
- Sourcing location
- Inventory levels
Then based on the scoring and the actual sales data, products which have the highest risk of stock-outs or the highest costs like transportation can be identified. This can help companies identify the units that would not be very profitable to the organization.
-
Dependency Identification
In the above point, we have spoken about “Supplier Risk” – this means identifying the number of suppliers that you are dependent on, and if any disruption at one supplier can affect your final product. This means you have to create a new plan, to identify the dependencies based on scoring each supplier.
This would allow you to minimize the risk and reduce the costs which have become important at this point where the shipping costs are at an all-time high.
-
Pursue Domain Transformation
Though use-cases are the way that companies try to solve their supply chain problems but think about it, the supply chain as we described at the start of this article is not a chain, but an inter-woven web consisting of multiple nodes, where a change in one node changes the other.
For example, take a look at the picture below- you can see how for a smart allocation strategy it is important to understand the links and to bring about transformational change, it’s best to pursue use cases within the same domain. Effectively, to scale up digital and analytics prioritizing fully enabled domain transformations is the best way.
Source: McKinsey
-
Merchandise Re-allocation
You must have seen Amazon delivery notifications that drill down even to the delivery warehouse where the goods are being transported to. This shows that the current machine learning algorithms have been developed so well that you can know in real-time whether any re-allocation is required for materials. For example, if you have a warehouse near the port, based on the delay in shipping you can decide to ship put parcels directly to the delivery centers.
-
Availability notification
Leverage real-time inventory alerts to not only re-allocation but to on-shelf availability, which can help in re-strategizing based on SKU availability. For example, when one SKU is sold out, ML algorithms can suggest substitution in real-time based on other inventory availability.
Other use cases in Data science for CPG Industry
-
AI to develop an adaptive view of the customer
If manufacturers and retailers can collaborate then they can leverage AI & ML algorithms better to be more adaptive. This can help with developing a complete view of the customer and leverage all channels of engagement, from interacting with online chatbots to sales associates in the store. This can help in visualizing foot traffic to shelf health and connecting all the dots in between.
-
Business Model revamp
Understanding the price elasticity or the competitive landscape can help with connecting marketing and sales like never before. This can also help in identifying newer requirements that can be useful for the overall efforts like Product affinity (with techniques like Market Basket Analysis), Customer segmentation, and Inventory Management
-
Forecasting and Optimization
As older modeling techniques fail to incorporate the wide range of resources available in current times, precise AI-based systems can predict the quantity of product that is required.
These are just a few examples of how supply chain analytics helps resolve issues. By now, you must have understood that the nodes of the supply chain through very complex, can be simplified or at least made more transparent with the right analytics tools.
In short, by working on Machine Learning algorithms in various aspects of the supply chain, you can help your company create Customer delight, deliver promises on time, and navigate across the ongoing supply chain issues. It is also important to note that data teams more than just some good data – they also require structure, automation, resources, capability, and a deployment strategy. And to make this happen companies need a good data partner who has both the technical and industry expertise to deal with the data and give the right solutions to the use cases with the macroscopic view in mind.
If you are looking for one such partner, then Polestar Solutions might be right for you as your partner of choice. If you are interested in knowing more about our capabilities in CPG Analytics, feel free to drop us a message or simply explore our resources.