Thanks to TMC Consultores' Intelligent Store Segmentation model, a client who is the leader in the food sector in Mexico, increased its sales and profitability in the traditional channel by 12%, generating a net sales increase of more than US$ 2MM in just one year, without alteration of any other variable.
Considering the complexity of this channel in a critical year in México, this result is considered a a resounding success
Background:
About two years ago, one of our clients expressed frustration with the lack of growth in the traditional channel, beyond historical trends. Although various initiatives had been implemented, such as new execution guidelines by store, adjustments in the route-to-market strategy, and a large number of promotions, they were not able to deliver the same results as in other channels. The main problem lay in the lack of differentiation in channel management, which was based on basic and imprecise segmentation, focused on store physical characteristics, sales volumes, and location.
Shopper behavior omission: segmentation models used up to that point completely ignored the Shopper's profile, mission and behavior by store. A generalized successful execution strategy was generated, without discriminating what was really happening in each store.
Key questions to understand the need for different segmentation:
In this situation, it is important to ask some questions to better understand the importance of buyer-based segmentation:
Would a store near a large school sell the same products as one near a hospital?
Is purchasing behavior similar in a store in a lower-class residential neighborhood compared to one in an industrial area?
Does the buyer buy the same when their main mission is to stock up, as when they are looking for products for immediate consumption?
Clearly, the answers to these questions are negative. Buyers adapt their purchasing behavior based on various factors, many of which are unknown to companies.
Each store has different profiles of buyers with predominant behaviors. It is essential to identify these profiles to better understand the needs and preferences of each segment.
Implemented solution:
Using the sell-in information by store from the last 12-24 months, we built a segmentation of stores by clustering using machine learning, with the aim of identifying groups of similar stores based on the mix of best-selling products over time.
The generated segments group stores that at first glance are difficult to understand their similarities or differences.
Examples:
The model classified traditional channel stores that were in the same residential neighborhood into different segments. When field validation was carried out, we observed that stores with a high share of individual-sized products ideal for immediate consumption tended to have work areas, educational institutes, hospitals, or others closer to them. However, stores with predominantly family-sized volumes were basically surrounded by residences.
The same dynamics were observed between stores with significant differences in the price segments of products sold. It was possible to determine its correlation with the demographic characteristics of the shoppers who buy in the store. We insisted this could happen within the same residential neighborhood.
Benefits
Increase your sales in the Traditional channel. The implementation of this segmentation based on the buyer allows a better definition of the route to market strategies, success picture (especially assortment and planograms), allocation of equipment (refrigerators, displays, etc.), and promotions, by specifically targeting the buyer objective and its predominant buying mission. In addition, it provides a deeper understanding of the needs and preferences of shoppers in different stores, which translates into continuous and sustainable growth.
On the other hand, it allows us to better understand the needs of our commercial channel partners, support them in improving the shopping experience, and finally in increasing their own sales volumes and business profitability.
Conclusion:
At TMC Consultores we make available to sales teams, Machine Learning technologies that allow them to make the most of the information available to segment stores and focus their initiatives and strategies on the target buyer and your buying mission. This generates rapid and continuous increases in sales. The data provided leaves no doubt: annual growth of 12% in sales volume and profitability. It is a real case that generated more than US$2MM in net sales in one year.
In a future article, we will talk about the challenges in the implementation of this type of segmentation model.
If you are interested in learning more about our store segmentation model and how to increase your sales, do not hesitate to contact us.
Written by Juan Manuel Domínguez R., CEO of TMC Commercial Consultants. For information about our consulting or training services in this area, write to us at contacto@tmcconsultores.com
Comments