In-store reality mining: quantifying and modelling the long-term behaviour of shoppers, measuring product interactions using sensors that capture the real buying experience
DianalyticsTM permits real-time shopper segmentation on the basis of characteristics, conduct and the factors that determine choices. DianalyticsTM is the first web-based platform that gives marketing and media managers a comprehensive set of indicators and metrics to scientifically measure in real-time the effectiveness of marketing and communication activities (in stores, malls, transport hubs, etc).
The available data makes it possible to follow the shopper's path from the entrance to the aisle and the final purchase. We can measure how many shoppers enter the store (store traffic), how many visit an aisle (aisle traffic), how many see a product (potential shoppers), how many interactions take place with the product (touched) and the number of purchase transactions (sales), as well as identifying who the actual shoppers are (actual shoppers).
A graph compares the traffic data collected in the aisles and in front of specific monitored shelves, where there are products (tests and benchmarks). Also indicated are the number of viewers, i.e. those who observe the shelf (broken down by product).
Our performance indicators make it possible to measure the efficiency of a category and of each individual product, not only in relation to sales performance, but also in relation to shopper interactions.
Correlation analysis indicates the relationship between different variables. In this case, the market share is positively correlated with the sales index (purchasing acts by shoppers) and, in part, with product attractiveness.
The Correlations graph is dynamic and flexible: the user can freely select the variables to be correlated each other. In this case the product' sales are correlated with the dwell, the attention and interaction time, as well as with the touched products and the potential shoppers. The correlation indexes are available for the product in test, the benchmarked one and the whole category. The product in test’ sales are highly related with the number of potential shoppers and the attention time. This means that the sell out of this product is positively influenced by the shopper's attention.
Sale Funnel: This section highlights everything that happens from the arrival of the shopper in the store, and makes it possible to compare the profile of the shopper between different products.
Brand Positioning: The graph shows the strategic product placement within the assortment, comparing them at different times and during different promotional activities.
This section shows, for each product, quality indicators and display quantities: linear cm, shelf levels, number of facings, sales and value per space.
Rulex has been integrated into DianalyticsTM to provide a powerful associative engine for data management and analysis. The accuracy and precision provide a number of features that enrich the analytics sets and insights of the platform.
Clustering: This feature allows you to create comparable clusters of shoppers according to their characteristics (gender, age, etc.) and purchasing behaviour. Find out more
Rules: This feature defines the rules that determine different purchasing levels (low, medium and high) for a category and for a particular brand. Find out more
Factor: What determines the purchase? Factors defines the variables that impact on the purchase of a category. Find out more
What if: what would happen if ...? By acquiring all the available data, this simulation module makes it possible, compared with a single factor (i.e. price), to estimate the sell out, given changes to the factor itself.
Prediction: given a sample model, it is possible to estimate sell-out results at a national level, carrying forward precision error and degree of relevance.
DianalyticsTM is also available on mobile devices.
The mobile application is aimed at Key account managers and merchandisers. It offers control and product benchmarking, as well as performance indicators on the move.
Rulex - Clustering
Clustering: This feature allows you to create comparable clusters of shoppers according to their characteristics (gender, age, etc.) and purchasing behaviour. In the examples the circles indicate, for the two selected products, their market share in each cluster.
Rulex - Rules
Rules: This feature defines the rules that determine different purchasing levels (low, medium and high) for a category and for a particular brand.
The example refers to a given category of products, and shows that under certain conditions (attention span of the segment consisting of adult women increased by 1.5 sec., and the placement of products on certain shelves and with a certain number of facings) a certain level of sell-out can be achieved.
Rulex - Factor
Factor: What determines the purchase? Factors defines the variables that impact on the purchase of a category.
