How do you rate in Retail Decision making?
I came across a compact sample of typical EDW reports utilized to track and support Retail decision making, when trawling across Teradata’s website (https://www.teradata.com/t/assets/0/206/280/fbff8440-b4c5-4211-a03a-c933460f35e0.pdf). It raises the question for many organisations as to whether the time spent deploying custom developed reporting and processes for their EDW could not be better served by moving to a purpose built solution such as provided by the leading EDW vendors
In times past a company would ‘lock in’ key learning and intellectual property through the development of complex custom designs in back end Retail analytics. Focusing on maximising Customer data allowed a Retailer to:
· Use market basket data to help guide promotional activity and planning.
· Manage vendors based on their importance to the business.
· Modify localized assortments by store to satisfy profitable customers.
· Review product sales and inventory channel.
· Monitor and proactively manage vendor performance.
Having developed such monitoring and reporting capabilities an organization was faced with the challenge of maintaining the designs and reports to keep the necessary levels of granularity and relevance, as well as reflecting any functional or performance improvements on the base platform. Such a custom design would, however, impose restrictions on smooth platform transition and may have missed any new developments from the supplier e.g. Purpose built retail process designs and reports. While these ‘native’ equivalents may have lacked the detail and insight of a Retailer’s own reporting, developments in the market have allowed experience with multiple customers to be incorporated by platform vendors (in a space such as EDW) into their core products. This has enriched the level of ‘out of the box’ reporting so that similar reports are now considered standard functionality. A sample of such reports used for Decision analysis, for example as deployed by Teradata EDW, are as follows:
· Merchandise Unit Performance (for a Product, All Locations)
· Flash Sales Report (a Location by Product Groups)
· Price Point Analysis (by Product Groups for all Locations)
· Lost Sales (Potential Sales) Analysis (by Products for all Locations, for a
· product group)
· Stock and Sales (by Products and by Locations)
· Sales and Profitability (by locations and by products)
· Trend Reporting (a Product Group by Weeks)
· Fast, Slow Sellers (by Items)
· Markdown Analysis (by Product Groups)
· Cluster Performance (by two Item Traits)
· Average per Store (by Products/Product Groups)
· Price Point (by Location, by Week, Crosstab)
· Vendor Sales and Stock (by Vendors, graph)
· Vendor Comparison Performance, Profitability (by Vendors)
· Rate of Sales
· Pre, During, Post Promo Sales and Margin (by Products/Product Groups)
· Product Affinity (by Affinity Products)
· Price Point Sensitivity Market Basket Analysis (by Price Points)
· Sales, Expense, and Labour (by Locations)
· Top 100 Customers (by Customers)
· Potential Lost Customers (by Customers)
· Cross Channel Performance (by Channels)
· Average Market Basket (MKB)
· Statistics (by Product/Product Group)
I thought this would be a nice ‘sanity check’ for an organization to firstly, compare and assess any ‘holes’ in current report coverage but secondly, to open a debate within the organization as to whether it is tracking the correct reporting metrics. As often happens, a tracked metric mandates a behaviour which allows it to be met. So while these metrics could be considered to be based upon a ‘system’ perspective, they do provide a benchmark for the ‘minimum set’ an organization might expect. And, referring back to the custom development path mentioned earlier, possibly an initial argument for investigating a change of solution towards exactly such an ‘off the shelf’ solution (if a significant amount of overlap is seen between current ‘custom’ reporting and these ‘standard’ product reports).
And while we are talking comparative studies, here is a look at some work SAP did on a “Future Retail Centre” (http://www.sdn.sap.com/irj/scn/index?rid=/library/uuid/f08ba07d-fe2b-2c10-5c89-b66f3e0c6dd3&overridelayout=true#4) . SAP Research built this “Future Retail Centre” in Regensdorf, Switzerland together with academic, technology and industry partners to showcase research developments and retail thought-leadership. This demonstrated a consumer-oriented walk-through together with a logistics environment and looking at tools for a typical retail headquarters.
The consumer scenario reflected unique shopping styles and followed a hurried “quick shopper” with a mobile phone shopping list. Using this device the shopper was guided via a map through the store and a two-touch mobile payment or scan utilised to facilitate the buying process. A more leisurely “weekend shopper” could gain access to product information through a personalized shopping cart and the immediate, RFID-based tallying of products chosen. An optic weight scale and point-of-interest advertising triggered by the customized shopping cart completed the customer-centric shopping experience (shown on the relevant ‘shelves’.
In the logistics scenario, a warehouse was set up with a labelling machine which applied RFID tags to the cases which are then used in processes such as picking of mixed pallets, packing and goods issue / goods receipt, all enhanced and optimized thanks to RFID. A new addition to the Future Retail Centre showed pallet movements with a forklift truck which had been enhanced to automatically read its position through RFID chips in the floor and thus reduce errors in scanning locations.Retail headquarters concerned with analysis and optimization of pricing could utilize a “Price Zone Optimization” algorithm which uses historic point of sales information to guide a retailer through setting up a new pricing strategy.
In the management of vending machines, capturing real-time inventory is the basis for accurate accounting as well as replenishment optimization. The Smart Vending application sent data of each transaction as it happens to the ERP system.
In addition, the Future Retail Centre can utilize Second Life as both a visualization and collaboration tool for Retail Management, as well as exploring its potential as an alternative shopping ‘channel’. ‘Food for thought’ and worth considering whether these really are ‘future’ concepts or whether they may be achievable in the medium term after all.
