A recent study conducted by MIT Technology Review and Google found that 60% of the companies surveyed are using big data analytics and machine learning to gain competitive advantage. These companies are looking for multiple competitive advantage returns as shown on the chart below:

Looking forward instead of backwards

Until recently most companies have searched through historical data stored in their systems of record to see if they can better understand and predict future behavior based on past behaviors and actions. This data is typically organized by products and their performance (sales, margins and profits) not by customer segments and their performance (adoption, utilization, and evangelization) and therefore not useful at predicting future behaviors and actions.

The emergence of systems of engagement through mobile applications and omni-channel distribution options is redefining the customer experience and becoming a new source of real time customer data. These new sources of both structured and unstructured data (from social media websites and online search/purchase log files) can now be mined to provide forward looking insights into customer preference, market trends and user adoption. See the recent blog from my brother Geoffrey Moore Digital Systems Maturity Model for a more detailed description of these changing events: https://www.linkedin.com/pulse/digital-systems-maturity-model-geoffrey-moore/

Expanding the data business applications footprint

Not only are the sources of data expanding but the scope and breadth of their different business applications is expanding as well. Simply put, if your company is not using data based analytics to improve its operating performance, its customer engagement skills, its employee productivity capabilities and its supply chain management processes you are at a distinct competitive disadvantage.

The ability to integrate massive, granular data sets with in-database analytics is enabling a whole new generation of business applications as shown from this EMC report below:

• Multi-Channel Attribution Analysis – attribute credit for sale across multiple marketing channels such as display ads, websites and key word searches.
• Customer Churn – predict the probability of customers’ attrition based on usage activities, support requests, payment patterns and the social impact of friends.
• Product Maintenance – predict equipment failures from embedded data devices based upon product usage, maintenance service records and product performance history.
• Clinical Trial Performance – model different drug outcomes based on clinical trials to understand treatment effectiveness.
• Yield Management, Merchandising Markdown Management and Price Optimization – build time-sensitive models to understand when and how much to increase or decrease prices given real time demand and supply conditions.  

Some current big data analytics use case examples:

Smart routing traffic data: By 2020 more than 70% of mobile phones will have GPS capability up from 20% in 2010. Current estimates for time and fuel savings from real time smart routing traffic data will be $500 billion by 2020.

Connected vehicle data: A recently completed report from Frost & Sullivan on “Data Monetization in Cars” said that if all the 200 connected car data points were monetized it would generate $33 billion in value. Today the report estimates that only 15% of this data is being monetized.

BMW has partnered with IBM to launch their own data brokering marketplace model called BMW CarData. They have equipped 8.5 million vehicles with built in telematics systems that monitor the car and driver’s habits and performance.

Speech analytics data: Southwest Airlines uses speech analytics tools to gain deeper and more meaningful information from live-recorded interactions between customers and their personnel. This tool has enabled Southwest to anticipate future customer needs and thereby deliver a higher quality customer experience.

Financial markets data:  JP Morgan recently partnered with data analytics startup Mosaic Smart Data to help its fixed-income sales and trading business become more profitable. Their fixed income revenues fell 27% in the three months ended in September and they deployed Mosaic’s smart data technology to help the bank’s fixed income teams “quickly make better informed decisions.”

Operations data: McDonald’s has equipped some of its stores with devices that gather operational data as they track customer interactions, traffic in stores and ordering patterns. They’re using this real-time data to model the impact of variations in menus, restaurant designs, employee training and productivity, as well as, sales.

New business data: A major transport company that plays an intermediary role in its customer’s value chain discovered it was collecting enormous amounts of data and information on global shipments. Sensing an opportunity, it created a new business unit that sells this data to companies who want to improve their business and economic forecasting analysis.

Getting started: Some keys for success

1. Know what kind of things you’re looking for to help you target the right data streams for analysis. Industry experts say that the biggest reason most companies don’t get the value and insights they want from their data is because they don’t have a clear picture of what they’re looking for. Weather.com hires “people who know how to query their data and tell a complete and accurate story of what the data is saying.” 
2. Focus on a prioritized set of desired business outcomes. Christina Clark, chief data officers at GE says that “often teams will fail because they are expected to address too many business demands at once, ultimately being stretched too thin to make a meaningful impact.” 
3. Breakdown data silos. Jeffry Nimeroff, CIO at Zeta Global says “every data silo creates a barrier between interconnections that can yield value. For example, think about a rich user profile either connected or disconnected from website activity data. The more data than can be interconnected the better, as those interconnections are where predictive power can be found.” 
4. Create good data hygiene. Building data analytical systems and processes that enforce quality is a major factor in extracting the maximum insights and value from your different data sources. Nimeroff says that ensuring repeatability of processes and auditability of results are critical success factors. He also says that deploying data quality tools including profiling, metadata management, cleansing, sourcing help ensure better results and outcomes. 
5. Recruit executive sponsorship for your analytics initiatives. This will insure that all your analytics initiatives are directly aligned with and in support of the company’s strategic business growth goals and critical business performance metrics.

Increasing the market value and operating performance of your company in the new digital world requires that you harness data analytics as a major contributor to your competitive success. Whether it’s using that data to get faster and better insights into what your customers want, or increasing your speed to market for new products and services; or improving the efficiency of your internal processes success will increasingly be defined by how well your company uses data as a competitive currency.

As always, I am interested in your comments, feedback and perspectives on the ideas put forth in this blog. Please e-mail them to me at pdmoore@woellc.com