CHALLENGE
How to optimize store space?
Assortment management has traditionally been the primary driver in grocery stores’ success. All the more so in small stores with limited shelf space and demand since the fit to local demand is even harder to reach.
Space management goes in parallel with assortment management. On macro-level, it is important to have the right allocation between categories to maximize financial performance.
When moving downstream towards micro-level, the overall category performance is affected by sub-category space allocation, along with product placement within the shelf modules and the placement of individual products.
A retail chain with small stores ranging from 100 to 400m2 faced a challenge in optimizing current space whilst maintaining the current amount of SKU’s. Furthermore, they wanted to avoid major category location changes.
First store to go through the uplift was the smallest store with 100m2 store space and 3000 SKU assortment.
SOLUTION
Predictive Analytics and AI to drive space optimization
Instead of treating assortment and space as separate entities as is the prevalent methodology, all data related to store operations were synchronized in Space Optimization solution:
- Space data to map factual macro- and micro-space
- Product master data to map individual products attributes – price, size dimensions, weight
- POS data with historical sales performance data by product
- Demographics data of each store’s catchment area to understand the local demand
At first, space data was used to map the factual store space down to shelf-level and visualize the store as Digital Twin. Each SKU’s shelf location was linked to SKU master data as well as sales, margins, product loss and inventory value data to gain insights on space productivity and the effects of shelf location to business KPI’s. POS data together with catchment area demographics data was used to discover the local demand.
Synchronization of all data created strong basis for predictive analytics and artificial intelligence. While AI is handling the heavy lifting, assortment manager is the one guiding it by giving rules: emphasis on revenues or profits, specific brands, price points, facings, stacking order. Or alternatively, letting the mathematical models automatically provide the best possible spaces.
The selected space and assortment scenario is automatically visualized in the Digital Twin – and turned into planograms that match the stores actual space for deployment purposes.
RESULTS
Store digitalization, Sales growth and margin uplift
Maintaining the current SKU count of 3000, the optimized space excluded 328 SKU’s that comprised both long-tail and less circulating products and added 616 new SKU’s to the assortment.
Without category location changes – as was the prerequisite – the optimized space delivered total sales growth of 13.62%, units sold grew by 5.37% leading to margin uplift of 13.71%.
13.62%
Sales growth
5.37%
Sales (pcs) growth
13.71%
Margin uplift
Besides these improved financial performance metrics, net working capital was optimized by the narrowing down of long-tail SKU’s.
What’s more, the optimized shelving levels minimized out-of-stocks that led to improved customer loyalty.
Indirect impact was the more efficient communication and deployment of new tactics and assortment throughout the organization. This was due to the Digital Twin that visualized the entire store and showed how the assortment was to be deployed down to shelf level and on the individual SKU.
