Letizia Affinito, Author


The global retail industry is experiencing an unparalleled crisis amid Covid-19. Consumers are staying home and reducing their spending while they face uncertainties linked to their health, wealth, and jobs.

This situation is hitting retailers hard, as U.S. and EU retail sales dropped, respectively, a historic 8.7% and 9% in March, and moving forward are expected to continue dropping. In a historical breakdown, retail spending in the United States dropped again in April, reaching a record 16.4% as people avoided restaurants, bars, stores, and malls – or because they were closed – during the coronavirus pandemic[1]. As a result, retailers are laying off staff and struggling to understand their options. They will have to restart and survive in a challenging market.

How can leaders minimize losses and find opportunities in the current scenario?

Existing dashboards are not of great help.

With such an unpredictable market, leaders should avoid making decisions based on pre-COVID-19 data and realize that existing dashboards are not of great help. In parallel, leaders should develop analytical models and tools centered on emerging behavior patterns. For example, Harvard Business Review illustrates how employing data science tools can aid in improving customer experience and operations: retailers can better figure out the products which are still in demand, the preferences for social distancing shopping, and study the emerging shifts in purchasing patterns[2].

New technologies are proliferating and becoming available to the mass market

Leaders should also use this year to upgrade data infrastructure. This involves redesigning the architecture of data collection and storage such that newly relevant data can be quickly mined for insights. Followed by reengineering predictive models using more-focused data sets; fixing glitches in website analytics, and tagging practices that hinder the ability to draw accurate conclusions from website data. Lastly, revisiting key performance indicators and scrutinizing each formula’s variables. Moreover, new technologies are proliferating and becoming available to the mass market. Indeed, the global retail analytics market[3] will reach USD 13.3 Bln by 2026 while exhibiting a promising CAGR of 19.2%. This is attributable to factors such as increasing implementation of modern technologies such as internet of things (IoT) and artificial intelligence (AI), among others[4].

The scarcest ingredient for the effective use of analytics is managers’ understanding of what is possible

While data infrastructure and new technologies are addressable with reasonable investments, the scarcest ingredient for the effective use of analytics is managers’ understanding of what is possible. Data, hardware, and software are available in droves, but human comprehension of the possibilities they enable is much less common. Given that problem, there is a great need for more education aiming at promoting analytical thinking to compete on analytics. Davenport presents three main stages of analytical thinking[5]:

  1. Framing the problem: one of the most critical parts of a good decision process which includes problem recognition, and review of previous findings. At this stage, it’s worth some serious thinking about who the stakeholders are for the analysis you plan to undertake, and how they’re feeling about the problem.
  2. Solving the problem: The importance of the “framing the problem” stage becomes evident when we hit the modeling stage. A model is built specifically to solve a specific problem. In modeling you will need to leave out the unnecessary and trivial details and isolate the important, the useful, and the crucial features that make a difference. At this stage model building involves using logic, experience, and previous findings to hypothesize your dependent variable (the one that you are trying to predict or explain) and the independent variables that will affect it). You will of course test your hypothesis later; that is what differentiates analytical thinking from less precise approaches to decision making like intuition.

The recognized problem is first organized through a modeling process into the critical variables, which then become data after measurement. The data you gather, of course, should be driven by the variables you have identified in the previous step.

Since data itself doesn’t tell us anything, we need to analyze it to decipher its meaning and relationships. Data analysis entails finding consistent patterns; in other words, the relationships among the variables that are embedded in the data.

  1. Communicating and acting on results: Communicating the results of your analysis to your stakeholders is the final stage in this three-stage, six-step framework, and it’s essential. Even if you have performed the other steps brilliantly, nothing good happens unless you make this step well.

The essence of this stage is describing the problem and the story behind it, the model, the data employed, and the relationships among the variables in the analysis.

When those relationships are identified, their meaning should be interpreted, stated, and presented relevant to the problem.

The clearer the results presentation, the more likely that the quantitative analysis will lead to decisions and actions.

It is crucial at this stage to focus on communicating insights not just facts and, using visually stimulating presentations of data in ways that add to the clarity of the results, would provide better chances to “sell” your insights and persuade your stakeholders.

Getting a grasp on these fundamentals won’t make you an analytics expert, but it will help you make more effective use of marketing analytics to support effective decisions and compete in the market. In today’s business world, not knowing about analytics can be dangerous to your company’s prosperity.

[1] WAHBA P. (2020), Retail sales fell a record 16.4% in April—the worst drop in history, Fortune, May 15, 2020

[2] Angel Evan and Amber Rivera (2020). Retailers Face a Data Deficit in the Wake of the Pandemic, Harvard Business Review, June 22, 2020

[3] The Retail Analytics Market is segmented by Solution (Software and Service), Deployment (Cloud and On-premise), Function (Customer Management, In-store Operation, Strategy and Planning), and Geography.

[4] WAHBA P. (2020), Retail sales plunge a record 16.4% in April—the worst drop in history, Fortune, May 15, 2020

[5] Thomas H. Davenport (2013), Keeping Up with the Quants: Your Guide to Understanding and Using Analytics, Harvard Business Review