Promotion drives at least 70% of sales volume for most FMCG categories; nailing the right promotional strategy is key to the success of a brand in any category. To provide competitive intelligence in relation to pricing strategy, brands use a range of data sources – one being evidence from quantitative consumer research.
Today, the best practice research approach for this type of work involves using trade-off behavioural experiments (or Choice Modelling) to ascertain the interplay between brand, price and promotion in shopper decision making. However, these types of research experiments often suffer from a disconnect between the deliverables that research agencies provide and the needs of commercial pricing teams – the key client stakeholders who are the ultimate users of the research outcomes.
FMCG Choice Modelling projects typically culminate in brand-level elasticities, which show the impact on market share of a brand changing its price.
- However, FMCG brands do not have a single price. Instead, they have a pricing strategy made up of different pricing mechanics at different frequencies that play out over an extended time period.
This means that a brand can’t change its price per se. Rather, it can vary the elements of its overall pricing strategy, and how these play out across an extended time period. Extending from this, the in-principle idea of the existence of a price elasticity in a market where no brand has a single price is, therefore, inherently flawed.
- Further to this, in practice a price change of a brand (say a week when it is on discount) needs to compete with pricing strategies from other SKUs in the market.
- As such, the impact of any pricing change is not constant, but in fact highly variable depending on the dynamics of pricing of competitor brands in the individual week in which the pricing change occurs.
This means that, in addition to being “theoretical”, the elasticities generated from traditional FMCG Choice Models systematically overestimate the price elasticity of brands, as they disregard the interaction effect with pricing of other competing brands in the market.
- This is not to say that the underlying technique of Choice Modelling is flawed – it is just that the way the results are being viewed fails to take into account the complexity of the market dynamics.
To make Choice Model results usable, consultants need to move from a static cross-sectional and point-in-time “week-based view” to a “time-based view”, considering how the pricing strategies of different brands overlay over time. This “time-based” approach permits the estimation of the impact of making changes to an overall pricing strategy, rather than just considering individual price level elasticities.
Nature has developed a means of achieving this using a Monte Carlo simulation technique as an overlay to the underlying Choice Model results. This process works by considering the underlying Choice Model as a “week-in-time” and then projects the results over a time-period based – allowing for the relative pricing strategies of each brand in the market.
To find out more, contact us at firstname.lastname@example.org or come to Peter Stuchbery’s presentation at the AMSRS conference in Melbourne. http://amsrsconference.com/sessions/6c/