7 Types of Promotion Analytics and What You Can Do with Them (Part 2)

This two-part Tip series was adapted from a set of DemandTec Viewpoints blog posts authored by Elizabeth Magill.

In Part 1 of this series, we discussed the first three types of promotion analytics: past performance, in-flight, and affinity. All three of these are pretty straightforward, and none require modeling the data.

We conclude here with four analyses requiring modeling of data which makes the analyses both predictive and much more powerful.

4.   Promotion response analysis.With this type of analysis, retailers can see the lift that various products or categories get from different promotional tactics or discounting mechanisms. For example, they can tell if beer—or even more specifically, Bud Light 6-pack bottles—should be discounted as percent-off or through multiples, and how much that offer might also benefit from being on ad or in a specific placement.

5.   Forecasting. Demand forecasting enables companies to predict the outcome of a range of promotional scenarios to see which might best meet their objectives. This can be very useful not only for your planning but to better negotiate terms with suppliers and to assure the right type of support from their merchandisers. Key considerations in the forecast include incorporating the interaction between products, i.e., cannibalization and halo effects. Retailers will want to make sure that the forecasting engine can take into account all the demand causals that will be relevant to their promotional scenarios, in other words, how different forms of discounting and promotion will affect consumer demand.

6.   Optimization. This analysis builds on forecasting, but automatically reviews multiple scenarios and identifies the optimal answer based on the goals that are defined, such as revenue, margin or volume. Promotion optimization can be more complex than price optimization as there are more variables at play, and hence more variables to optimize. It's not about simply optimizing a price point, but optimizing against specific tactics, such as whether the offer is on ad or display. 

7.   Customer segment analyses.All of the above analyses are traditionally done at the category level, however, by using customer identifiable transaction log data, they can be conducted at the customer segment level, as well. For example, with an understanding of promotion response by customer segment, retailers can view how sensitive different groups of customers are to various promotional tactics. For example, does their Urban Singles segment respond better to multiples or percent-off for beer? Likewise, with the ability to forecast by segment, they can understand the potential trade-offs. They may be willing to give up margin at the category level, for example, if they know that they'll gain significant volume with their Urban Singles shoppers whom they have identified as an opportunity segment.

The importance of having these promotion analytics insights when managing the promotional process can't be overstated in today's competitive environment. Whether managing the promotional deals from manufacturing partners or promotional offers and events within the organization, these tools offer immediate, context-driven access to the analytics needed to make winning decisions.

LEARN MORE ABOUT THE BENEFITS OF PROMOTION ANALYTICS:

Submit this form for an instant download of the new DemandTec eBook: Promotion Collaboration - Five Steps to Success... or to request other information from DemandTec. We also welcome your inquiries and feedback on the topics discussed in this Tip.