In retail, marketing can create a visit without converting intent when the offer is unavailable at the right store, t…
In retail, marketing can create a visit without converting intent when the offer is unavailable at the right store, time or variant. Measurement must connect exposure, catchment, traffic, availability, conversion and contribution. A national stock or revenue average can fund campaigns in zones unable to serve demand and attribute an execution problem to media.
Decision and method.
Activate or reinforce local campaign only in stores where expected availability converts incremental traffic with positive net contribution. Build a store-week table for traffic, conversion, effective availability, net price, margin, local spend and events; document closures, works and promotions. Estimate incremental traffic with experiment, control zones or quasi-experiment; otherwise label attributed traffic as a journey indicator, not causal proof. Multiply incremental traffic by eligible conversion, availability and unit margin. Test cost and timing of stock correction before raising media. Use quadrants: amplify high-availability/high-return stores, replenish proven-but-constrained demand, diagnose weak conversion and temporarily exclude unserviceable locations.
Worked example.
A €30,000 local campaign brings 5,000 incremental visits. Eligible conversion is 20%, margin €60, effective availability 82%. A €4,000 stock plan raises availability to 96%. Without correction: 5,000 × 20% × 82% × 60 − 30,000 = €19,200. With correction: 5,000 × 20% × 96% × 60 − 30,000 − 4,000 = €23,600. Differential gain is €4,400. Stock precedes media growth; activate only where availability can exceed 88.7%, the break-even threshold for the €4,000 correction.
Checks and limits.
Align exposure, traffic and availability by store-week; measure availability from customer viewpoint, not only system stock; separate incremental from merely attributed traffic; include stock-correction cost before media decision. Availability can be endogenous to local demand expectations. Cross-channel spillovers need suitable window and geography, and average improvement can hide permanently absent sizes, colours or references.
Resources and sources.
Download the retail-availability dataset. Related: measure omnichannel journeys, calculate full channel cost, size reach. Sources: Campo, Gijsbrechts and Nisol (2000); Fisher, Gallino and Xu (2019); Verhoef, Kannan and Inman (2015).