For Revenue Growth Management (RGM) teams, the challenge lies in the interdependence of portfolio decisions. A price increase on one pack size can shift demand to another. A new product launch can cannibalize existing SKUs before attracting competitor volume. A delisting that looks profitable in isolation can expose volume to competing brands.
Traditionally, portfolio profitability optimization meant reviewing each product’s P&L in isolation. This works for a single product, but it can't capture how shopper behavior shifts across two or more products or portfolio and category-level effects such as cannibalization.
Portfolio profitability is not only about which products are profitable today, but it also includes which combination of variants, prices, pack sizes, and distribution best protects or grows portfolio contribution while maintaining category position.
Modern RGM teams are moving from SKU-level margin analysis to portfolio-wide simulation. Using wargaming to optimize product portfolio profitability allows teams to test different portfolio configurations before decisions go live and understand how shoppers, SKUs, retailers, and competitors interact across the full category.
At Buynomics, we believe agent-based simulation is the best approach to optimizing portfolio profitability. Buynomics’ Virtual Shoppers AI simulates individual shopper decisions across your product range and competitive set, helping teams model where volume is expected to go when a product’s price changes, a new SKU is introduced, or an existing product is delisted.
Buynomics' Virtual Shoppers AI is grounded in behavioral economics and decision theory. The model is trained on real-world data, including transactional data, surveys, and conjoint studies, to create 100,000s Virtual Shoppers that replicate how your shoppers make purchasing decisions while accounting for market context, competitive dynamics, and cross-category effects.
RGM teams define the hypothesis they want to test, whether it's a change to their pricing, PPA, promotions, mix, or trade terms, or a competitor action, such as a new product launch, price cut, or increased promotional spend. Both can be modeled directly, allowing teams to evaluate their own next move and anticipate how the market will respond to a competitor's. During each simulation, these virtual shoppers make purchasing decisions through the following process:
The model breaks down products into key value drivers and attributes that influence purchasing decisions.
.svg){kind=icon}
The model analyzes the data and identifies the willingness-to-pay (WTP) distribution for product value drivers.
Through data updates, it continuously learns and adapts to different market factors.
Each time you run a “what if…” simulation in the Buynomics platform, virtual shoppers are making decisions holistically - accounting for all RGM levers, competitors, and trends/seasonality (incl. special purchasing moments, e.g., Christmas).
.svg){kind=icon}
Accurate shopper behavior prediction results in 2-4%* profit impact.
When presented with product offers, Virtual Shoppers can replicate actual buying patterns with 95%*accuracy.
Reduce the time spent on manual analysis by up to 80%. Integrate various data sources and get best-in-class predictive accuracy.
