Weldom
Targeting a population that changes with the seasons
Catchment areas resized from real mobility flows, budgets reallocated store by store, impact isolated by comparison with control stores: how Mobsuccess made the local targeting of a neighbourhood retailer as shifting as the local areas it serves.
The context
Weldom holds a distinctive position in French retail: that of the neighbourhood retailer rooted in rural areas and on secondary traffic hubs. Its stores serve local areas on the edge of the pull of large cities, and face two structural difficulties.
First, significant retail leakage: rural populations naturally turn to neighbouring cities during their purchase journey.
Second, demographic instability: a significant share of the stores are located in holiday areas, where the population present and the size of the catchment area vary sharply from one season to the next. Targeting models based on a fixed radius in kilometres were no longer capturing these real flows.
What we did
Mobsuccess built a dynamic geomarketing set-up for Weldom, in four parts. Territory modelling: the brand’s CRM data is merged with GPS mobility signals to resize catchment areas continuously and estimate the population actually present at any given moment, distinguishing permanent residents from seasonal visitors.
Win-back targeting: journey analysis identifies the competing brands that customers turn to once they leave the area of influence, and concentrates delivery on these leakage points, at the moment of the purchase journey.
Local management: budgets and formats (display, social media, CTV) are reallocated store by store according to conquest potential and seasonality, with messages adapted to the type of customers present. Incremental measurement: performance is isolated by GeoLift, comparing activated stores with unexposed control stores.
Results
Compared with unexposed control stores, activated stores show an incremental return on investment of +66% and a cost per incremental visit cut in half.
- Catchment areas continuously resized from real mobility flows, rather than from a theoretical radius in kilometres.
- Performance isolated by comparison with control stores, separating the footfall generated by the campaign from natural footfall.