The use of AI for retail players can be a decisive factor in their business recovery
Despite an increase in ecommerce activity since the quarantine measure was introduced, retailers, placed in a state of economic emergency, are facing sharp drops in sales. In a global context of cost control, marketing today is under pressure to quickly generate traffic and in-store sales, while reducing costs. The contribution of Artificial Intelligence (AI) in these processes is now a crucial lever to maintain commercial efficiency while considerably increasing productivity.
As a result of the crisis, store closures, and the difficulty in securing workstations in supply chains have led to a drop in sales in stores and ecommerce.
With no sales or cash flow, brands have no choice but to act on two priority performance levers. Retailers must first reduce their operating costs in the short and medium term and tightly control their working capital requirements.
As a result, marketing is likely to be one of the first collateral economic victims of this crisis with an obligation to significantly adjust all spending (media, agencies, platforms, etc…) and a priority focused on commercial efficiency.
To do this, the brands will have to arbitrate and choose to allocate their resources where they will have a rapid and sustainable return on investment.
The so-called Pareto’s Law suggests that 20% of the causes are responsible for 80% of the effects. This theory is obviously a misleading simplification, and may even lead to strategic errors such as focusing loyalty only on very good customers.
However, this theory reminds us that for a marketing campaign all suspects, i.e. all the recipients of an operation, do not have the same potential. Obviously, depending on the objective of the operation (promotions, clearance, novelties, etc…), the right suspects are not always the same.
The challenge for marketing teams will, therefore, be to identify suspects that can potentially generate a higher ROI according to a precise objective, in order to significantly reduce mailing costs while maintaining a high level of performance.
This is where predictive marketing comes into play. Exploiting customer data (offline and online), predictive models now make it possible to quickly detect the best suspects according to a marketing objective.
Predictive marketing is revolutionary in that it statistically predicts the best suspect customers for a campaign. Marketers will then be able to focus only on “ROI operations”, with ROI being simply the difference between the fixed and variable costs of a campaign and the incremental revenue generated by that campaign.
On the one hand, we drastically reduce volumes and therefore expenses, on the other hand, we maintain performance, a real contribution to the service of marketers.
