Overwhelmingly, 68% of retailers cited “data organization” as their top data-management challenge. Following that figure are the 65% that said they struggle with data integration while 43% ran into some roadblocks with data capturing. Data access and data storage, struggles for 30% and 28% of retailers, respectively, were lesser concerns. One of the greatest changes since 2012 was the fact that managing data volume, the top concern last year, is no longer at the top of the heap.
Retailers may, in fact, trail other industry verticals in terms of adoption of big data analysis tools and techniques because of the nature of their businesses. Agarwal noted that some industries, particularly financial services and pharmaceuticals, are in a seemingly better position to run advanced analysis of high-volume data than the retail industry.
Compliance and business risk within these heavily regulated industries, Agarwal said, necessitate substantial investment in big data analytics. For instance, these tools are used to “help banks identify and prevent fraudulent transactions, [which is] more directly tied to the brass tacks of their business model than to the idea that big data can help improve customer engagement in retail.”
While retailers may be comparatively slower to adopt analytical solutions, this vertical excels in developing analytics solutions themselves in-house either by building or acquiring. Major retailers like Target and Walmart are pouring what Agarwal described as “hundreds of millions of dollars” into big data “exploration.” Moreover, Staples acquired personalized offers startup Runa while Home Depot snapped up pricing engine BlackLocus.
Although large, progressive retailers are making the data investment, there is a "dual reality" that exists, where, on average, a $1 billion retailer invested less than $75,000 on big data experimentation this year, the bulk of which will go toward proof of concepts and ROI-mapping to determine future scale, EKN noted. This indicates the industry has much room for growth in reaching data maturity.