The Next Frontier of Data Analytics

The inaugural wave of customer-oriented data analytics focused on digital marketing and improving the customer experience. Why?

First, the feedback is really fast: an enterprise knows how, when, and why targeted consumers respond to a given marketing action. Given holistic data capture, the enterprise understands how a product or service was used, the nature of the customer experience, and the likelihood of a repeat purchase.

Second, an enterprise can precisely measure how effective certain actions or representations are from an ROI perspective. With techniques like A/B split testing and machine-based learning, data-driven marketing actions can be quickly tuned for optimal results. As a consequence, sales, marketing and customer-targeting optimization have been the predominant areas of analytics investment.

This quick feedback loop is akin to a sugar rush. But there can be real energy in that sugar. One major hotel chain that used data analytics with real-time data to answer questions about customer preferences increased annual booking revenues by ten of millions through dynamic room pricing optimization.

The next frontier in data analytics is operations efficiency. The benefits can bigger, broader, and potentially more sustainable. Examples of applying analytics to operations include:

  • Processes, decisions and advice undertaken by Robo capabilities
  • Just-in-time inventory management (e.g., parts stocked precisely to meet demand or 3D printing initiated to produce needed parts)
  • Work scheduled, load-balanced, and directed to resources or entities best able and available to execute
  • Proactive cost-efficient maintenance (e.g., plants, vehicles, aircraft or other machines that communicate their mechanical status and enable maintenance scheduling accordingly)
  • Dynamically priced services and materials to minimize waste or underutilized capacity
  • Anti Money Laundering (AML) and Know Your Customer (KYC) intelligence to reduce false-positives
  • Surveillance using visual machine learning (e.g., high-definition scene recognition or object detection in public spaces)

All positive, but the above benefits can be harder to achieve than marketing-related data analytics. Applying data analytics to operations and day-to-day decision-making requires changing processes, technology, and talent at a fundamental level. To reach this frontier, an organization needs new ways of sourcing, ingesting, processing, and storing data. To pursue such strategic ambitions typically involves five major steps:

  1. Identify the operational areas and customer-experience processes where significant improvements could be realized with more, or better, data.
  2. Prioritize the next-generation or digital-alternative operating arrangements where data could make the greatest difference.
  3. Determine the data requirements and what it would take meet these requirements across existing and new/alternate data sources.
  4. Develop a case for action and confirm where and how this can be verified with more analytics, prototyping, and select pilots.
  5. Plan and prepare an operational blueprint; change strategy and performance metrics associated with the envisioned operational transformation and data-related capabilities.

Today, leading firms are casting a broad net in terms of collecting data—from social media telematics, advanced sensors, video and IoT (Internet of Things) technological capabilities. The shift from using only structured internal data to also using unstructured internal and external data is transformative. Integrating data from multiple sources enables a step change in operational efficiency that has the potential to transform business models. In the end, data-driven operational efficiency is much more impactful and sustainable than data analytics for marketing. It’s worth exploring this new frontier.