Can Data Save Us All? Lessons from Attempting to Transform Sears and Kmart through Data & Analytics
Lessons from attempting to transform Sears and Kmart through data & analytics.
Disclaimer: I worked at Sears Holdings from 2013 to 2015.
Data and analytics are often heralded for their ability to transform business – making companies smarter, faster, more collaborative, and better able to serve customers. Yet from my experience working in targeted marketing and data analytics at Sears Holdings, I know that data’s promise isn’t always delivered, and I wanted to offer some lessons from my experience.
To set the context – Sears Holdings launched its ShopYourWay rewards program in 2009 as a customer loyalty program designed to facilitate customer purchases across online and in-store channels and across retail brands such as Kmart, Sears, and Lands’ End.[1] By the time I joined the company in 2013, ShopYourWay members represented 65% of total sales pof over $36 billion in the fiscal year ended 2014 and were involved in over 100 million transactions each year.[2] In the four year history of ShopYourWay, we had amassed a unique data set, tagging individual people to specific SKU shopping behavior (not merely total basket data) gathered from purchases both online and in store.
Where We Succeeded
Using data to communicate with shoppers more effectively and remind them of our value proposition – Driving from segment insights to more granular individual-level insights
One of the most successful programs we implemented was a targeted marketing campaign around bringing back shoppers who exhibited a change in their frequency of shop. This “member reactivation” campaign generated millions of dollars in weekly incremental transactions (as measured against a control group). From there, we constantly monitored shopping behavior to manage the profitability of individual members and maintain their shopping behavior. Furthermore, we moved from a baseline of a single customer segment of customers who hadn’t shopped in a few months to a RFM (recency, frequency, and monetary value of shop)-driven model with over 100 customer segments with targeted offers to each segment based on the behavior we hoped to promote.
Where We Failed
Our data strategy did not align with our corporate structure
When I joined the company, ShopYourWay’s data-driven marketing was being hailed as the center of the company’s transformation efforts. The fiscal 2012 10-K stated Sears was “transitioning to a [ShopYourWay] Membership company”.[3] However, our internal organization was not organized with data and analytics at its core – instead we had a matrix structure of business units representing categories (e.g. home appliance and grocery), selling channels (e.g. online and store operations), and data operations (e.g. ShopYourWay).
Most importantly, ShopYourWay member data was owned within a distinct business unit which provided analytic services to the other business units, complete with internal charge structures that impacted business unit P&Ls. The motivation was for the ShopYourWay analytics team to prove their value to the organization, but the outcome was an acrimonious relationship between the data science team and the operational business units. It would have been far better to share the data openly given data and analytics were meant to be at the core of the company strategy.
The purchase data we used reflected a subset of the entire customer experience, missing data undermined our delivery of a complete customer value proposition
Our data analytics team mostly focused on transactional data and product-related behavior such as online browse paths. However, delivery of a full retail business value proposition depends on creating an experience for the customer, not just understanding the products they buy/consider buying. Our corporate team should have augmented our transactional data with qualitative data from focus groups and surveys to understand customer motives for shopping our owned channels, as this may have helped us address key issues in our business model. Instead we simply focused our data efforts on creating better merchandising and promotion, and perhaps missed out on opportunities for us to support other important elements of the customer experience based on data insights.
Conclusion
At Sears Holdings, our focus on customer data allowed us to drive significant results for the business. Ultimately though, we were limited by the scope of our data and the organization we built to execute on it.
Going forward, I will remember that data and analytics are only as good as the latest insights generated from them, and that thinking flexibly and involving an organization fully in the culture of data creates the best opportunity for analytics to drive a business forward.
[1] http://comparerewards.com/archives/1123
[2] http://www.chicagobusiness.com/article/20130823/BLOGS10/130829907/another-thing-that-hasnt-saved-sears-loyal-shoppers
[3] Sears Holdings 10-K for period ended February 2, 2013
Awesome post, Alex. I think there are some very valuable lessons in here. You mention two points of failure, but of those, I would argue only one was actually controllable — embedding analytics more fully into the corporate structure. Do you think the second failure you mentioned could have been avoided? In hindsight, it’s easy to see what you missed, but you can’t measure everything. Ultimately, you’ve got to identify variables you think matter and hope they will continue to matter into the future.
Thanks, Alex! The question of how to organize a data analytics team at a large company like Sears is an interesting one. Given the importance of data / analytics to drive strategy and operational decisions, should companies have a c-suite officer dedicated to analytics / data? Some level of company-wide oversight and management seems especially important given the importance of cybersecurity threats and data compliance. I’d be interested to learn more about how other companies have done this well, but perhaps there could be a sort of federated model in which analytics talent reports to one person centrally but is also embedded in local units.