Category management is a strategic approach to retail marketing and a set of methods that enable retailers to go beyond the industrial or sales-based logic of selecting, presenting and promoting their offer, by integrating the specific needs of customers, such as their in-store purchasing behaviour or product usage logic, in order to build lasting customer loyalty. For brands, category management means supporting retailers in the development of their categories by providing them with market expertise and action plans tailored to their performance and strategy. https://www.impp.fr https://www.linkedin.com/in/bernard-bourdon-phd-5017aa10/

Merchandising, as it is understood in the context of retailing, consists of the manner, strategy and technical means of highlighting and making products, brands and promotions visible to shoppers in stores. The term "e-merchandising" also applies to online stores (e-commerce sites).

The consumer is the person who consumes the product for his or her own use. The shopper is the person who makes the purchase (the buyer) in a physical store or a digital one (e-commerce). The shopper may or may not be the final consumer of the product. A common example used to distinguish these two terms is the purchase of pet food. The shopper is not the consumer of the product. Although some retail professionals sometimes seem to blur these two terms together, in practice we have seen a real differentiation between these two "populations". In fact, when we observe the behavior of a shopper in a point-of-sale who makes a purchase in a matter of seconds, strongly influenced by a colorful environment presenting a plethora of brands and references, we obtain results quite different from when we question the same person as a consumer at home on how they perceive the way they buy a product. We believe it is important and necessary to introduce different research methodologies to deal with the question of the consumer and the shopper. In this respect, our own research at Pomelo enables our customers to obtain very different data and results today than with other consumer-oriented research methodologies.

Behavioral tracking is the technique used to measure shoppers' purchasing behavior when shopping in a physical store or e-commerce site. By behavioral tracking we mean, in particular, the shopping journey, stops in front of products on the shelves, product pick-up and purchase. This technique makes use of a variety of tools, from tracking sensors, to video analysis to wifi or Bluetooth tracking systems for cell phones, or so-called neuromarketing tools.

Eye-tracking is often confused with behavioral tracking, whether in-store or online. Eye-tracking, as the name suggests, involves tracking the eyes of participants to understand where and how they look at their environment. In general, this is a qualitative approach, which focuses on how people read a planogram (the organization of products within a department) or look at communication and visual messages. What's more, eye-tracking measurement bases are small, as it's a cumbersome technology to implement (each participant has to be recruited individually). What’s more, it remains difficult to quantify visual behaviors and link them directly to purchase conversion rates. Eye-Tracking shows its strength when working on communication and search strategies. Behavioral tracking broadens the scope of measurement by making it possible to quantify behaviors in a more general way, and to link them much more easily to purchase actions and conversion. It is therefore a more immediately operational and effective approach from a commercial point of view. Eye-tracking can complement behavioral tracking to enhance understanding, particularly in terms of visibility and visual appeal (packaging, point-of-sale communication, etc.).

This term comes from a diamond-shaped area of visibility in the center of a shelf, where shoppers (usually measured through Eye-tracking devices) tend to focus their attention. This result comes from a number of studies carried out mainly in the laboratory, in which subjects are placed motionless in front of a real or virtual shelf (projected onto a screen). This result has often been transposed to real life as a description of the area of greatest visibility of a shelf (a diamond in the center of a shelf). However, real-life eye-tracking using wearable technologies has shown that this result is hardly representative of the visual attention paid to the shelf by a shopper while shopping in a real, cluttered, crowded environment. It remains valid, however, in the case of shelves located behind a cash register, which shoppers face in a stationary manner (behind the cash register in a tobacco shop or pharmacy, for example).

When it comes to collecting shopper data at point-of-sale, we often tend to validate the relevance of a methodology by the number of stores included in a test cluster. However, this depends entirely on what is being measured and what is expected of it. Most of the time methodologies that take clusters of stores into account are called quantitative, and are aimed at measuring sales figures. To do this, we select a group of stores and average the results of sellout data, in order to assess sales trends over time. To be representative of a large market (on a national scale, for example), we'll need to select a large number of stores (often several dozen) to obtain a reliable estimate of sales trends over the whole territory in all its diversity. It's important to note that the only information generally available is the number and value of products sold in the various categories, without any knowledge of the traffic in the aisles, the transformation rate, the purchasing path, the visibility of products, etc. This is also the reason why it's so important to be able to estimate the evolution of sales in a given territory. This is also why we need to average sales over a large number of stores, in an attempt to limit the effect of these unknowns if we were simply comparing one store to another (there may be significant disparities in traffic from one store to another, or even simply in the visitor rate of a category due to its location in the store with equivalent overall traffic). Other methodologies, such as in-store questionnaires, will require fewer points of sale, because when a population is correctly sampled in a few stores (4 or 5) by questioning a few hundred customers per outlet, we learn no less than if we carried out the same exercise in several dozen of stores. Behavioral tracking methodologies take a similar approach: they’re able to precisely measure traffic, purchasing paths, hot and cold zones, conversion rates, the effects of promotions on behavior (stopping power), and here too, if we are able to analyze this for several hundred or thousands of people per store, there's no need to multiply the number of outlets tested by dozens. What counts is the diversity of the types of layout for the categories we want to test, but we learn no less about a layout when we test 2 or 3 stores than we do when we test 20 or 30 with the same layout. To sum up, we can classify the different methodologies according to the average number of stores to be tested and the results expected: Quantitative approach (sales figures in particular, but loyalty card data can also be included): at least a dozen stores per location to be tested, or even several dozen when you want to know the evolution of a market PROS: precise, large-scale measurement for a clear understanding of market trends, or for measuring sales trends between different category (or store) locations. CONS: provides little insight into the obstacles and levers to the act of buying between different category (or store) locations, and does not provide a clear understanding of how to improve conversion rates and visits to different categories and segments. Qualitative approach (in-store questionnaires): a few stores per region, or possibly per category location. PROS: understanding how consumers/shoppers perceive a category, their needs, expectations and difficulties, as well as their uses and habits, in order to better deploy the offer or work on its breadth and depth. CONS: it's also difficult to assess the obstacles and levers to action, as it's difficult for consumers/shoppers to rationalize their behaviors, and there's a lot of perception bias, which makes declarative information imperfect for guaranteeing the effectiveness of the merchandising implementation that could be deduced from it. Behavioral approach (aisle traffic, conversion rates, buying paths, stops and grabs, purchase acts, product visibility, etc.): 2 or 3 stores per layout tested PROS: precise measurement of the effectiveness of various category merchandising layouts, enabling us to deduce the optimal organization for sales growth CONS: does not reflect market trends or large-scale sales figures, and does not tell us how consumers/shoppers perceive a category.