Evaluating Store Customer Experience
43-store mystery shopping data identifies the four stores that need help.
A retail chain with 43 stores across 7 regions wanted to know which stores actually need help, and where to focus management time. The instinct is to check every store. The data shows that's not necessary.
Mystery shopping data was collected across the full customer journey — booking through checkout — using a consistent question set so stores could be compared fairly across regions.
The overall network average was 72.9%. George Town and Melaka stood out at 88.7%, with all stores in those locations performing consistently well. Most other regions sat broadly within range. The bigger concern wasn't average performance — it was the spread within "Other Locations," where store scores ranged from 93.3% down to 32.0%.
Only 4 stores (9% of the network) qualified as true outliers, with gaps greater than 15 points below their regional average. The most severe scored 32.0%, sitting 38 points below benchmark.
Statistical testing showed no meaningful regional clustering of underperformance (p = 0.807). One pattern cut across all regions: greeting and engagement was weak in 34.9% of stores — a network-wide behavior issue, not a regional one.
Three customer journey stages drove ~20% of the total performance gap: product selection, promotion explanation, and booking. Most of the severe gaps were behavioral (75% in critical stores), which means coaching, not process redesign.
The action: stabilize the four outlier stores immediately, run one network-wide greeting fix, leave the rest alone.