Modeling a Shopper's Decision-Making Process

empty

Modeling a Shopper's Decision-Making Process

Best-in-class providers must delight customers, not simply satisfy them. The complication for a best-in-class retailer is that it must delight consumers with different shopping styles and different objectives. In fact, depending upon the occasion, the same consumer might be a moving target. For example, during one visit, the consumer's objective might be to quickly pick up a missing ingredient to a recipe; on another visit the objective might be to shop for a holiday dinner.

Complicating matters further, store features that might delight a shopper looking for convenience may conflict with features that would delight a person looking for variety or novelty. To be best-in-class and dominate competitors, a retailer must delight all shoppers, not just a segment, and not just sometimes, but everyone, always, and without compromise. By modeling the shopper's decision-making process, retailers can identify store features that will accomplish this objective.

What is the best approach for painting a portrait of how consumers shop and evaluate options at a retailer? The standard approach--which is correct in most cases--would be to conduct a regression-based driver analysis to help identify which attributes and features drive consumers' satisfaction and purchase interest. The problem with this type of analysis is that only one driver amongst several that are very similar to one another will be chosen to represent all of them. This results in a model that, at best, provides an incomplete understanding and, at worst, no understanding at all of what drives a decision.

Even if this regression problem could be solved, the model may still not be the best one. For certain types of retailers, a driver analysis may fall short because (1) consumer purchase decisions vary markedly depending on occasions and personal involvement, and/or (2) purchase decisions may be non-compensatory; that is, consumers will not make trade-offs between attributes, such as trading quality for a lower price. In these situations, the recommended approach would be a decision model and a penalty-reward analysis.

Decision Model A decision model would

  • determine both the importance of individual attributes as well as the optimal combination of attributes that generate positive ratings for satisfaction and purchase interest;
  • evaluate the `ideal' combinations of attributes/retailer features that would appeal to different segments of consumers (e.g., involved buyers versus uninvolved; novelty seekers vs. basic shoppers) and how occasions (e.g., just a quick stop or a more focused effort targeted at finding hard to find items) may change customer expectations;
  • be used to develop a scorecard that can be used to evaluate best-in-class performance.

Penalty-Reward Analysis A penalty-reward analysis will identify those attributes/features that delight customers in order to

  • target where the biggest satisfaction gains (rewards) would be realized (e.g., with product improvements)
  • target where the smallest satisfaction losses (penalties) would be suffered (e.g., with cost reductions, lack of attention, etc.).

The following penalty-reward analysis identifies those attributes that will delight customers (rewards) or detract from satisfaction (penalties).

In the example above, quality of merchandise has the potential to create both a penalty and reward. That is, there is an increase in overall satisfaction for having quality, but there is a larger decrease in overall satisfaction for having poor quality. That will be the penalty that costs customer loyalty. Attributes like courtesy only serve to reduce overall satisfaction. Attributes like value offer the opportunity to increase overall satisfaction. The recommendations based on the above might include:

  • Fix attributes that offer a significant penalty. These may be considered as a price of entry as there is no reward for doing it right, but doing it incorrectly detracts from overall satisfaction and costs customer loyalty.
  • Leverage attributes that offer only a reward to add value to the consumer. If value is added, customer loyalty will increase, and delighted customers will emerge.
  • Prevent dissatisfaction and increase overall satisfaction with attributes that offer both a penalty and reward.

The research described here will not only lead to precise results, but actionable results. Input from this research can be used to drive changes in strategy and execution to attain a dominant retail position. It will help the retailer to achieve best-in-class status, meaning that it will delight customers, not simply satisfy them.

More insights about Public Sector

Society