Point of View

Analytics in times of uncertainty

Using augmented intelligence for more informed business decisions

  • Facebook
  • Twitter
  • Linkedin
  • Email
Explore

During times of uncertainty, enterprises across the globe face unexpected challenges. To manage these challenges, protect business continuity, and build resilience, governments and businesses must make big decisions – and fast.

However, it's critical to balance judgement and emotions with data-driven insights to optimize decision-making. This is the premise of augmented intelligence.

Analytics teams can help their businesses unlock data-driven insights by focusing on three areas:

  • Immediate transparency and visibility
  • Insights for targeted actions
  • Continuous monitoring

1. Immediate transparency and visibility

That enterprises understand how a global business challenge is evolving – on a daily, or perhaps even hourly basis – is a critical and immediate first step. Most enterprises create cross-functional response teams – or nerve centers – that empower managers and subject matter experts to respond to a situation with creative and pragmatic solutions at speed and scale.

Figure 1: Composition of an integrated nerve center – Source: McKinsey

View

To be successful, nerve centers need transparency and daily reporting across these areas:

  • Employee safety and business continuity
  • Financial top line and working capital
  • Distress in operations and implications for key products, suppliers, and customers
  • Customer base and customer sentiment, especially for B2C industries

Only with this transparency can analytics teams make informed recommendations for the best next step.

Take a copy for yourself

Download PDF

Case study

Visibility for the media and entertainment industry
Depending on the nature of the global challenge, some businesses will experience a bigger impact than others. For example, in response to COVID-19, a large media and entertainment company decided to close all its properties, which had a huge impact on its revenue. To manage this closure and prepare for its eventual reopening, its analytics team had to work in new ways.

Previously, the analytics team analyzed key performance indicators (KPIs) related to revenue, market segmentation, and inventory optimization. Instead, the team created KPIs – based on customer sentiment and customer query data – to analyze customer cancellation patterns and customer behavior. This daily reporting and reprioritization helped the organization to review and adapt business plans and customer communications accordingly.

2. Insights for targeted actions

For enterprises to make decisions with confidence, they must analyze multiple scenarios before they act – even when pressed for time. Unfortunately, changing directives and advice issued by different countries can create fragmented workforces and place serious constraints on the ability to act.

This situation calls for decision makers and industry experts that can devise practical solutions while remaining flexible. Actions taken during a period of disruption will reduce the immediate impact on customers and supply chains while building resilience against future shocks.

To stay focused on creating feasible alternatives, analytics teams should run simulations and stress-test various options using business data. This helps leaders make decisions by considering the probabilities of various scenarios and the pros and cons of each.

Figure 2: Cluster maps reveal alternative sourcing options for all materials affected – Source: McKinsey

View

Case study

Building supply chain resilience with analytics
Advanced analytics is particularly helpful for manufacturers who must constantly review supply chains during periods of disruption. Facing plant closures and restricted transport options, Genpact helped a global manufacturer who wanted to make informed operational decisions regarding its stock-keeping units (SKU), inventory, manufacturing lines, and suppliers.

Today, a rapid analytics pod is helping the supply chain business leader with the following:

  • Machine learning-driven forecasts: at SKU level based on a category risk index, customer behavior heat maps and insights into new consumption patterns
  • Smart inventory optimization: at store level based on an early warning system that considers risk levels across regions, days of inventory on hand (DOH) across suppliers, transport limitations, and the trade-off between stocking at distribution centers versus direct shipping to retailers
  • Alternative sourcing recommendations: by performing multi-tier risk optimization models that also consider risk levels across regions and DOH information

3. Continuous monitoring

To improve business resilience, enterprises must protect baseline operations and develop an insight-based action plan. In addition – and for what is perhaps the most critical step for continuity – the enterprise must continuously monitor the progress of the situation in question and make necessary adjustments to the action plan.

Analytics leaders should set up situation rooms where they can monitor business performance on an hourly or daily basis (depending on the nature of the business) to ensure that operations are always on track. If possible, video conference capabilities can also speed up unified decision-making.

Case study

Under the pressures of COVID-19, a regional bank wanted a real-time view of lending risk across its consumer and small business portfolio.

To help, the analytics team deployed an early warning system to continuously monitor customer activity. They are also monitoring shifts in spending patterns and uncovering insights into merchant data. This allows the bank to spot risks, act, and even anticipate future challenges.

Rapid analytics pods

In time of uncertainty, speed trumps repeatability and pragmatism trumps sophistication. Most traditional ways of working will not work in this environment. We recommend that analytics teams be set up as rapid analytics pods to support the integrated nerve centers and business leaders.

The pods will need multidisciplinary skills including data engineering, data visualization expertise, data science, and business analysis with industry expertise. These pods should work on a hyper-agile delivery model with 24-48-hour sprint cycles. Plus, most analytics teams will have reusable codes and assets including data models, quality and variable transformation routines, visualization templates, and simulation model structures that they can repurpose.

If there's one positive to uncertainty, it's that businesses have an opportunity to experiment in an environment where there are fewer barriers to change. Uncertain times can create interesting opportunities.

Learn more about how Genpact can help with augmented intelligence.

Read more insights to support your business' response to COVID-19

Learn more