Customer Feedback Instore Testing
Understand the drivers for customer satisfaction, when analyzing test and control groups of own stores and local competitors
Customer: Global Retail Company
Challenge: Testing new store concepts for a group of stores
Objective: Monitor customer feedback of test stores vs. control groups vs. local competition
Solution: Dashboarding and NLP Analysis of Google Reviews for certain topics. Comparison between test group of stores vs. control groups, as well as selected competitors in neighborhood
What were the key challenges?
One of the country organization was going through a major rehaul, as it was competing against very mature competing chains.
The integrated customer project team, was testing a new instore concept for 10 pilot stores, ranging from changes in staffing, selection as service at checkout.
One key challenge was, how to measure quickly the difference in customer perception between test stores vs. the remaining stores in the country. In addition, the project group was keen on understanding, how customer would rate the pilot stores vs. selected competitors in the neighborhood.
How does our solution differ?
Rather than incentivizing customers to fill in structured survey data, we used the NLP platform of our MIGO suite to analyze Google reviews of pilot vs. control vs. competing stores.
The advantage of this approach is, that customer feedback is unbiased, as google nudges Android users to fill in a short review , when leaving the store.
Sourcing all historical reviews as well as daily sourcing of new reviews, the project team was able to analyze customer feedback from to 500k reviews. Hereby, the project team was able to understand the topics of interest as well as the sentiment (positive, negative, neutral)
Predict42 set up its automatic sourcing routines for our own as well as relevant competitor data within days. By applying a special designed topic model, we are now analyzing and acting on any changes in customer feedback between stores in test and pilot group.”