Using Power Map to Analyze Public Retailer Data


Here is a video that I put together for Retail showcasing Power Map with public data available for Nordstrom, Bloomingdale, Saks & Neiman Marcus. This is part of a visualization that we put together to showcase several layers of public data available for retailers. Taking the public data and combining with proprietary data can enable retailers to get a complete picture about their customers and market. This combined insight from internal and external data enables retailers to better cater to customer's needs, personalize the experience and better predict what customers want based on trends, correlations and patterns.

I began with Twitter data that I acquired using the Excel Twitter Addin. I acquired twitter data for the last 7 days for some of Nordstroms brands: Nordstrom, Nordstrom Rack and their online entity Haute Look. I  took the geo data from the tweets and plotted them on a map to give me insights into locations where specific brands are popular. This can be very useful for a retailer to discover areas in the country where there is significant brand activity. When correlated with store locations, search and other data, this can be even more powerful.

I also pulled in Bing Search data for the same brands and analyzed search traffic around the country. The shading for the states show where specific brands are popular. In this visualization, I overlaid the historic search trends over the years from Bing on the more real time twitter data to show where brand activity was historically vs over the last few days.

Yet another layer of data that I made use of is Google Trends data to compare search traffic for Nordstrom, Bloomingdale, Nordstrom Rack, Saks and Neiman Marcus. This layer in the visualization shows the areas in the country that have a higher affinity to specific brands. This is an excellent way to determine areas that represent opportunity for growth.

Data from Yelp enabled me to obtain information about stores in a particular city, especially for analyzing location data for the stores and doing a proximity analysis for competition. I used an API call to Yelp (no coding required) using Power Query and acquired store data (including location and Yelp rating) in New York for Nordstrom, Saks, Bloomingdales and Neiman Marcus. Using these layers of data, Retailers can also overlay actual store performance data and compare stores visually, correlate with search, twitter, yelp and demographic data.

The final layer that I pulled in was US Census data. This was a data set that I imported into Excel with detailed demographic information by Zip Code. With this layer I can correlate store performance by Median Age, Median Income, Population and so on. For this visualization, I used the Median Income data and displayed it as heat maps. Using this, I can correlate areas of high and low income and compare store performance data against this.

Here are a few more examples:





Comments (0)

Skip to main content