The CDO’s Role in Crisis Management
Over the last several months, I’ve had the opportunity to be involved in discussions with chief data officers (CDOs) about how they are dealing with the impact of the COVID-19 pandemic. Across industries, there are three key themes that have consistently surfaced.
Dealing with the Initial Shock
A crisis such as a global pandemic creates an uncertain and highly volatile environment. CDOs I’ve been in discussions with talk about the need to quickly assess conditions, develop immediate response plans, and prioritize actions. This requires acceleration of data curation and use, which many cited as a challenge because of manual processes for data acquisition, cataloging, cleansing, and access. However, the good news is that many CDOs saw a significant increase in the breaking down of silos and in the level of collaboration between internal IT and business teams, as well as between government agencies and cross-industry organizations. Most teams met at least once a day to review data, identify hot spots, and respond, accomplishing things in days or weeks that would previously have taken weeks or months prior to the advent of COVID-19.
For example, a healthcare provider recognized stock planning that was tuned to historical demand parameters was no longer valid. They quickly re-evaluated safety stock levels and adjusted planning parameters across individual hospitals and the network of hospitals they managed, in order to make data-driven decisions regarding the allocation of inventory across the network. They also prioritized item replenishment with suppliers to mitigate stock outages and improve health outcomes. The onboarding of a new supplier—which previously would have taken a couple of weeks—was accomplished in a couple of days.
Minimizing Operational Disruption
Once past the initial shock wave, the CDOs turned their focus to ensuring business continuity and building operational agility. Common actions here include helping finance manage liquidity and cash flow through strict discipline around costs and collecting receivables, as well as helping manage and consolidate financial and operational data to support forecasting and scenario modeling. The CDOs were also asked to provide data to support ongoing monitoring, analysis, and communication with customers, employees, suppliers, creditors, investors, and regulatory authorities.
For example, an international logistics company that specializes in beer, wine, and distilled spirits analyzed outstanding purchase orders and prioritized collections, which helped accelerate cash flow by millions of dollars. They also realized that when restaurants, bars, and venues were closed due to COVID-19-related public health measures, kegs of beer would begin to expire. They started a new service to decant or remove the beer from the kegs that were past their shelf life and return the empty kegs to the breweries. This was a completely new business opportunity that they had never thought of previously and required new data to be successful. The process included being able to demonstrate to local authorities that the beer disposal complied with all environmental laws, and providing a complete audit trail of which kegs were decanted so that customers could get a refund of duties paid on the beer from tax authorities.
Planning for Recovery
Finally, all of the CDOs talked about looking ahead and incorporating lessons learned into their scenario models for forecasting demand, predicting variability in supply chains, determining ways to safely bring workers back into offices, and figuring out how to stay connected and engaged with the organizations and people they serve. As part of these discussions, they also talked about the need to re-examine their models, the assumptions that went into building the models, the data sets used to train AI/ML algorithms, and the bias created by data about previous performance on the predictive accuracy of future outcomes.
For example, a travel organization talked about incorporating more external data to broaden the context of decision models for return-to-service planning. The completeness and accuracy of factors that impact revenue and costs—such as demand, pricing, cancellation, and load data—were critical for obtaining valid insights. They also looked for ways to change and automate business process to increase efficiency and productivity, such as by combining IoT data from engines with external weather data to improve fuel efficiency.
While the COVID-19 pandemic has created uncertainty and volatility, it has also brought people, teams, and organizations together. By breaking down silos and using data collaboratively to deal with the initial shock, minimize operational disruption, and plan for recovery, companies across industries have been able to effectively manage through the crisis. Although the discussions were specific to the pandemic, the methods and analysis used are applicable to any kind of crisis.
Learn more about steps that data and analytics leaders can take to adapt to the COVID-19 crisis at www.informatica.com/crisis-response.