Using structural equations modeling to understand store performance drivers

Structural equations modeling is an extremely good technique to model multiple cause and effect simultaneously. Take for example we want to trace the causal chain from foot fall to conversion to spend dispersion to monthly sales turnover to store profitability. How do we get to see the complete cause and effect chain. Techniques like regression etc cannot handle one outcome variable recursively being a causal variable

Clustering stores

Stores can be clustered on the basis of various store performance metrics and differential treatment strategies can be adopted. This could be with respect to the

1) Store format to be adopted for each store cluster

2) Mix of merchandize to be stocked for each store cluster

3) Pricing strategy for each store cluster

4) Promotion mix strategy for each store cluster. Coupons vs Temporary price reduction vs Gift off

5) Instore experience strategy . Lighting vs staff per square feet vs Displays vs sampling

6) Shelf placement strategy

Stores can be clustered on a variety of attributes like

1) Price sensitivity to strategic products

2) Footfall in store

3) Spend dispersion observed in basket across categories like grocery, hair care, electronics, juices etc

4) % revenue accrued from loyalty card holders vs anonymous buyers

10 behavorial variables to segment Retail shopper

Here are some possible variables to segment customers using their behavorial profile

1) Customer Purchase Recency
The number of days which have elapsed since the customer last purchased from a store
Example : less than 30 days, less than 3 months, less than 6 months etc

2)Tenure
The number of months the customer has been a member of the loyalty card program
Some targeted campaigns could have increased loyalty subscriptions during certain periods

3)Average basket value
Average amount spent by the shopper during each visit to the store
Example : less than $ 50, 50à125 $, greater than 125 $

4)Average basket size
The number of items the shopper purchases during each visit

5)Spend dispersion
The % of spend dispersed across various categories of products like music, books, stationery items, perfumes etc
Example : 12 % on stationery, 35 % on books, 43 % on perfumes

6)Customer Purchase frequency
The number of times the customer purchases from the store in a year
Example : 10 times in a year

7)Overall spend dispersion profile
The % of deviation between spend of this customer and an average store shopper to benchmark the intensity of shopping on various categories relative to an average buyer
Example: An average Joe would spend lets say 20 % on stationery, 50 % on books and 30 % on perfumes. If David’s spend dispersion profile is 60 % on stationery, 35 % on books and 5 % on perfumes, his spend bias helps us understand his profile better relative to an average Joe.

8)Demographic spend dispersion
The % of deviation between spend of this customer and the demographic segment to which the customer belongs to

9)Range of products purchased
Out of the overall number of categories present in the store, what % of the categories has the customer purchased

10)Range of channels used
A product can be sold thru multiple channels – company owned store, franchisee store, web , phone/contact center

What are 5 important levers a retailer activate to stimulate growth to shoppers basket item count ?

One of the basick checks in Retail analytics one can do by processing POS Transactional data is to verify if the size of the basket in terms of number of items shopped for in one visit to the store is increasing or decreasing. Once we find that, the trend in basket count is coming down when benchmarked with no of items in an average shoppers basket size one year back or even a quater back then we can activate 5 possible levers.

Action trigger-1 : ACTIVATE STORE LEVEL BUNDLED PROMOTIONS
Can we create a new store promotion to entice shoppers to buy more ?Can we have bundled promotions. Example if you buy a soap and a shampoo you get 2 % off the shampoo. It could be a 'gift off' ( where a consumer gets an item free with the purchase of another item ) or temporary price reduction scheme or weekend coupon promotion scheme is working well . In case of food items, can we have live sampling for food items in store which consumer normally does not purchase ? How about an SMS blast to people living within the store vicinity ?
Action trigger-2 : CHANGE MERCHANDISE MIX AFTER SURVEYING SHOPPERS
It is conceivable that consumers come to our store and do not find specific items which they would like to purchase. A simple survey can be administered to shoppers to find out "What is it that you would like to see in our store ?". If there are common items which are occuring frequently in the response then the merchandise mix can be changed in the store to include the most frequently requested item by shoppers which is currently not displayed in our shelves.
Action trigger-3 : CHANGE SHELF PLACEMENT AND VISIBILITY STRATEGY
In some cases it has been found that even though the SKU ( Stock keeping unit) is present in the store, shoppers do not purchase it because it is 'buried' deep in some corner shelf and not easily visible to the eye. A simple planogram multivariate data analysis can reveal co-relations between shelf placement ( At eye level, Below eye level, Above eye level, near POS counters ) and its effect on the items purchased by a shopper during a visit.
Action trigger-4 : INTRODUCE LOYALTY CARD AND HAVE BEHAVIOR BASED RECOMMENDATION ENGINE
In order to track a customers specific behavior in terms of visit frequency, average spend, dispersion of spend across multiple product categories, recency of visit etc it is important that a loyalty card program be instituted. Once this is done we can use the torrent of purchase behavior transactions to create a behavior based recommendation engine. Techniques like collaborative filtering and market basket analysis can be used to derive cross sell recommendations. Once this engine gets trained on past purchase behavior it can be used to derive the top 5 recommendations for each of the loyalty card customer based on past behavior exhibited. This surgical approach has a greater chance of stimulating increase in basket count as opposed to 'carpet bombing' them with a generic recommendation on product
The competitive stores within the immediate 3-4 mile vicinity of our store are offering the same items at a higher cost. Shoppers find our store competitively priced for the items they shop .
Action trigger-5 : Incentivise staff based on customer recommendation and increasing store staff per 100 square feet of store space
There have been scenarios un earthed in stores where a customer comes to the storeand cannot find knowledgeable staff who can guide her on finding certain products . This problem can be alleviated by dynamically increasing the number of staff dynamically based on footfall observed during different shifts, ensuring store staff are sensitized to customers needs and incentivising store staff who have been recommended by shoppers who found the staff courteous and helpful

Action trigger-6 : NEGOTIATE VOLUME DISCOUNTS WITH VENDORS
If a competitor store in the vicinity offers SKU's at a price point below our store, then there is an opportunity to show past purchase and POS volume for the vendors SKU and negotiate a volume based discount which ensures that our store is competitive in terms of price for an item displayed resulting in the shopper purchasing the SKU and ensuring no dip to average basket count from our stores

Depending upon the dynamic nature of the market and competitive activities, one or more of the above levers can be used in conjunction to ensure that as a store manager one is able to manage a upward trend in the average no of items purchased per shopping visit

Store metric - Shelf alignment index

Definition of shelf alignment index

Shelf alignement index is a simple metric which stores can use to check if the merchandise displayed on their shelves is in alignment with shopper behavior at the store. For example lets say

Input data required to calculate shelf alignment index

1. POS transactions ( SKU, Store, date, um, qty, value )
2. Shelf space audit ( Category, date, shelf area in sq ft, relative share of shelf )

How does one calculate shelf alignment index

1. Rank all categories ( soaps , shampoos , vegetable, ice creams ) based on the relative number of units sold per month.

2. Rank all categories on share of shelf

3. Calculate difference in rank = Rank(Sales)-Rank(Share of shelf)

4.
Example of shelf alignment index


Application of shelf alignment index to optimize store operations