Leveraging advanced analytics to make proactive decisions.

In years past, store traffic was a simple measure: the number of shoppers who entered the store and traffic counts were critical in determining core retail metrics like conversion (i.e.; the percentage of shoppers who actually purchased and calculates as the number of transactions divided by store traffic).

Traffic only told the story of “how many.” Retailers wanted to know more about their shoppers, but additional data collection included methods like surveying, and was very manual and very expensive.

When customers visit the stores, retailers try to convert the traffic into sales profitably through several means. They invest in store labor to ensure that customers experience a good shopping service that encourages customers to purchase and return to the store in the future.

  1. First, predicting walk-in and gaining an understanding of how traffic affects store performance facilitates the development of effective labor planning and scheduling models and the effective utilization of store labor, which is the second largest expense for retailers. This understanding further helps retailers identify their “periods of potential,” their key selling periods, and optimize their service levels not only by assigning the right number of associates to the selling floor but also having the best performing associates work during those key periods.
  2. Second, predictive walk-in allows retailers to identify appropriate key performance indices to benchmark performance of different stores. Traditional store performance metrics such as sales and profits do not provide the whole picture, as they do not reveal to retailers the sales potential of their stores and the ability of their stores to convert that potential into sales.
  3. Finally, the ability to forecast traffic and evaluate its impact on store performance can enable retailers to identify strategies to increase sales for each store and lead to better coordination between marketing activities and labor planning/scheduling activities.

Our model analyze the statistical properties of retail traffic patterns and build predictive models that could facilitate store traffic planning as well as scheduling.

Store managers might allocate FTEs at a fixed rate over the week basing their choice on a non-objective knowledge of the walk-in weekly pattern.

The SDG model allows FTEs allocation relies on a statistical prediction of the walk-in. Then the optimization model allocates the FTEs as best as possible.

The store walk-in is also an important KPI to analyze the effect that a store window design has on sales.

You can obtain a prediction of your future revenues using our Walk-In Prediction,
data on conversion rate and median invoices. Hence you can choose your budget more reliably.

You can match your products and the skills of your sales staff to the customer’s needs using further information about the Walk-In: age, language, previous purchases, etc.