Tapping the Untapped Market : The Sequel
In my previous blog, I had talked about how weather can influence a retailer’s data governance model. Here we will explore more factors that can spice up the same.
Holiday/ Special event Forecasts
Consumers remain enthusiastic about Holiday Shopping -both in-store and online. Studies show that consumer demand for seasonal products can increase 50 to 500 percent during the holidays. So is the case with special events like Super Bowl, World Cup etc. But what we forget is that sales for such events start at least a month in advance of the event itself. Also holiday and special events have both obvious and subtle sales trends. Jersey sales may pick up a month or even two before Superbowl but the sale of beer will increase only during the days of the game. By optimizing the inventory flow to the stores for seemingly “subtle” time periods, retailers can steal a march on the competition and ring up some additional sales.
Historical data can be used to calculate holiday/ special event effects on store traffic, seasonal inventory levels and customer market basket mixes. So it is very important to segregate the categories that needs to be dynamically governed during specific time periods. To tackle this, retailers need to capture the same data points that we had discussed earlier under Weatherization i.e.
- Seasonal item category information: Retailers have to identify a means to categorize the information to various seasonal buckets e.g. add a keyword to the specific items. This is to ensure that one or many categories that spans across the merchandise can be selected for a particular event.
- Geography data: Country-> State-> City data as certain weather trends can be geography specific.
- Sales:Past sales trends to identify the categories that need to be classified as seasonal.
- Lift to be applied to a category:Lift defines the percentage of increase in items to be pushed to the stores to meet the demand. E.g. It was discovered that hosepipe sales increased as it became warmer, but only to a certain point. So the percentage of increase should be adjusted using lift.
- Time period:Ability to choose the time period for which the rules should be applied.
Third party data
Today’s retailers rely on third-party data to get a full 360-view of a consumer. Third-party data is data that retailers can act on from providers like Nielsen, MasterCard Advisors, Equifax, eXelate etc. Third-party data providers capture information directly from first-party sites or transactional data from other retailers which is cleansed and compiled to make the data effective and usable. With third party data, retailers will have access to the complete information about their customers and there will be minimal risk of missing key behaviors of their likely audience. Third party data can help retailers capture changing demographics, lifestyle trends, spending habits etc. which can again influence decisions around what, where and when. This data also transforms the way businesses carry out customer segmentation. Instead of limiting segmentation to internal fields such as “geographic location” or “products purchased together”, retailers can achieve more strategic segmentation based on external data such as industry size.
Again, to handle this form of dynamicity, retailers need to understand and use data points like Item classification, Geography data, Lift to be applied to a category and Time period.
News events can drastically shrink the response window for retailers and will require the products to be on the shelf during a short period of time. For example, a handbag retailer’s sales may rise because a celebrity appeared on the news with one of its products. Other such examples include major life events in a celebrity’s life, new movie releases, political events etc. When the musical legend Micheal Jackson died, from the day of his death i.e. June 25th until June 30th, U.S. retailers of physical CDs had to order about 3 million copies of his Sony Music Entertainment albums as opposed to their normal order quantity of about 45,000 copies. In such scenarios, retailers should have an option to push such in-demand items alone to the customers in a jiffy. The caveat here is that there may not be any historical sales data to determine how much to push to the stores. Retailers should use their “retail intelligence” to formulate a number to be applied as lift to specific items within a category for a specific time period.