Consumer Goods companies are seeking to leverage advanced analytics, machine learning, and artificial intelligence to gain a competitive edge as the industry continues to be hammered by disruption, changing consumer buying habits, and proliferation of indie brands.
In search of hidden sales and market share opportunities, companies are discovering the benefits of Revenue Growth Management and leveraging the analytics that tackle three critical challenges for CPGs.
- How to determine the right product mix and pack offerings to improve sales?
- How to determine the right list price and promotions to offer to customers by channel?
- How to determine the right timing of product placement, price changes, and promotions?
To overcome these challenges, CPG companies deploy strategies to change the customers’ perception of their product’s value equation on the shelf, while influencing the consumers’ purchasing decisions with increased product relevancy and loyalty.
Best practices in Revenue Growth Management requires a disciplined analytical approach to predicting consumer behavior; and accurately aligning product prices, placement and availability within each channel by understanding the impacts of core demand drivers that increase growth and profitability.
Let’s evaluate how advanced analytics can solve two challenges encountered by Consumer Goods companies: hyper-competitive and multi-channel market.
- Establishing the right list price to grow market share
- Reducing trade spend and over-promoting
Establishing the right list price has traditionally been based on intuition using past actions and negotiation leverage. While the negotiation leverage never goes away, data-driven companies use regression models to assist with pricing decisions. This often leads to looking at price elasticity alone which lends itself to becoming reactionary to competitor price moves in the market or gravitating to raising prices for inelastic products only.
Advanced analytics provides a better method for understanding how to establish a pricing strategy to gain market share and protect profitability. One analytical approach to achieve pricing optimization is to use an ensemble of models. Many companies capture pricing decisions for both their products and competitors through third-parties, such as IRI and Nielsen, which can be used in these models. With this transactional (consumption) data, you can create an ensemble approach to address:
- Attribution models to cluster products based upon crucial attributes that help define the consumer’s purchasing decision tree for your products compared to similar competitive products by defining choice sets (substitute products that have a high relational impact).
- Attraction models to define the behavior of products within a choice set reflecting what purchasing decisions that can occur at the shelf.
- Linear-Mixed models to then be applied to optimize the pricing recommendations by applying the right business constraints and objectives to maximize market share, revenue, sales growth, and margin.
Reducing trade spends and over-promoting is another area within Trade Marketing and Sales where advanced analytics can improve efficiencies significantly. Traditionally, trade spend can account for approximately 30% of revenues and is one of the largest P/L items for any Consumer Goods company.
Understanding how to distribute this trade spend by understanding the evaluating drivers of demand is paramount. This evaluation provides insights into when, where, and what will drive uplift over the base business and drive incremental growth and profitability. Often, companies repeat promotional history and react to competition based on intuition, while customers (retailers) demand more and more via “pay to play” with trade term spend to remain on the shelf.
With advanced analytics, companies can address repetitive trade promotion problems by using a myriad of codified best practices and then use an ensemble approach to solve the problem. The results are increased trade execution and profitability.
- Bayesian Model: building a better baseline at the SKU/store/week level is critical for success to understand the factors that drive demand above the base where promotions, holiday, outliers, and other effects are removed.
- Random Forests: can be applied to rank, thus determining the most significant attributes that make promotions effective.
- Linear-Mixed Models: are applied to optimize the baseline forecast by anticipating each promotion attribute, such as promotion pack, price, holiday, and promo offers, that serves as a driver of demand.
Overall, these are just a few examples of how advanced analytics can solve trade promotion and pricing problems for Consumer Goods companies. Advances in artificial intelligence will continue to offer new methods for using behavioral science. This knowledge enables companies to improve internal processes from collected data derived from e-commerce channels, resulting in a greater understanding of consumer demand.
To see other ways companies are unlocking analytics in the CPG industry, see the article from Boston Consulting Group entitled, Unlocking Growth in CPG with AI and Advanced Analytics, published October 15, 2018.