A regional confectionary chain with 200 stores in major urban areas across two European countries. It had been recently been acquired by a private equity from a retail conglomerate company.
Growth had stagnated for the client, which is why the parent company had decided to sell the chain despite it having strong brand recognition. The new owners wanted to answer two key questions:
How should new stores be selected
What are the key parameters for ongoing store portfolio management.
The project comprised three stages: (a) compiling store data, (b) developing the analytical model and (c) converting the analysis into implementation steps
The compilation stage consisted of collecting locational factors that included type of store, location (e.g. mall/high street), accessibility for each unit along with others. Additionally, operational factors such as staffing levels, hours of operation, territory managers were collected. Financial performance data at a unit level was then used to enrich the data.
Development of the model had three major components. First, the high performing stores were separated from the low performers. A list of discriminating factors were created and then a weightage was assigned to each factor. Lastly, the factors were segregated into two groups depending upon whether the impact on the factors is largely driven by location or by store operations. The data was then processed in the model.
Some of the key insights were that the correlation between profitability and rent was weak despite a positive relationship between sales and rent. Another insight was that advertising spend seemed to have a positive impact upon driving store traffic. Clustering the stores into three different groups also enabled the identification of operational factors that clearly separated high and low performers and could potentially reverse the performance of many of the low performers.
The work enabled a fact-based prioritization of units for refurbishment while identifying key discriminating factors for selecting new store sites. A pilot group of low performers is being used to test the operational levers that have an efficient yet positive impact upon performance.