Is travel in your blood? This headline in the newspaper caught my attention last week. The article talked about a low-cost airline having partnered with Dr. Richard Paul Ebstein, a world-renowned expert in genetics, human biology and neuroscience, to determine whether there is such a thing as a wanderlust gene. If you carry the 7-repeat or the 2-repeat of the dopamine D4 receptor gene, DRD4, you are part of 20 percent of the population that is more inclined to seek out novelty, travel and adventure.
Travelers have been invited to join an experiment to have their saliva tested to earn the badge of someone with travel as an innate need, in the process enabling the airline to gain insights into their travel preferences based on multiple variables (background, cultural preferences, opportunities, to name a few) and send customized travel-related information and offers their way. In other words, the airline and the expert are currently in the process of collecting data (lots of it) before they can analyze it to make business decisions for an organization.
In a separate situation, I recently met a potential data science candidate for a recruiting conversation, and as we were getting coffee, I casually asked him whether he sees most of the data science community tending to come from a particular field of study or experience. He (equally casually) responded by sharing with me the analytics results of his trawl of global LinkedIn profiles.
Analytics is here to stay. Multiple sources place the market size for analytics at approximately $200-250 billion by 2025, with the field categorized into four segments: descriptive and diagnostic analytics, which look to the past to explore what happened and explain why; and the futuristic categories are predictive analytics, which aims to forecast what will happen (based on historic data), and prescriptive analytics, which explores how we can make it happen.
How different sectors benefit from analytics
Automotive provides numerous use cases in manufacturing (near real-time demand analytics in collaboration with dealerships for postponement of customization), quality management (product defect analytics to provide early warnings of failure for recall planning, remote performance tracking to predict spare parts demand) and parts availability (Inbound 2 Manufacturing analytics to optimize scheduling of inbound parts ordering and delivery to reduce component inventory).
Consumer & Retail features prominently in any analytics-related conversation. Product and channel analytics help define omnichannel strategies and reduce product launch lifecycles, while consumer analytics is what we saw in our wanderlust example as well. Multiple retailers now apply sentiment analysis to consumer feedback, resulting in more accurate forecasts for production planning, reduced time to market, minimizing obsolescence and hence reduced need for discounting.
Energy & Chemicals has been leveraging analytics in upstream and midstream for a long time, with newer use cases now being identified in downstream and renewables. Overall use cases for the sector in upstream include reservoir interpretation, optimizing low-rate wells, matching real-time drilling data with production data from nearby wells, doing intelligent pipeline monitoring, in downstream this includes understanding distributor demand for chemicals or lubricants, and discussing how convenience retail can become as key a business line as fuel. For renewables forecasting failures in solar plant sensors and predicting weather conditions to make wind energy projects bankable are key use cases.
Engineering & Manufacturing has use cases to increase efficiency, especially in manufacturing. These include risk modeling to predict sourcing delays, develop contingency plans and diversify the supplier base; equipment failure prediction and pre-emptive maintenance; and production line analytics to reduce raw material and work-in-progress inventory in multi-stage manufacturing.
Getting the edge on the competition
Life Sciences and Healthcare has use cases in business model innovation, with patient analytics for on-demand healthcare and R&D analytics for reduced batch sizes and inventory levels; and in terms of shipment tracing and analytics of data from multiple sensors to ensure product integrity in cold chain logistics and eliminate the risk of counterfeit products, which can prove fatal to lives. As the sector explores increased direct-to-market operating setups, analytics provides the speed and flexibility such supply chains will require.
In the Technology sector, demand forecasting is a prime candidate for product launch support, especially for consumer technology products with short life cycles and hype-based demand. Leveraging analytics to optimize the inventory of spare parts for any capital equipment, as well as predictive maintenance for increased uptime and predicting aftermarket demand, can provide just that extra edge over a competitor.
What organizations need to ask themselves
Supply chain & logistics is embedded in all the above, with sector-agnostics implications for the industry. Diagnostics and descriptive analytics can be applied to current supply chain flows to identify optimization potential, e.g. in consolidation, warehouse footprint reduction, transport optimization, risk and resilience planning to identify key facility locations in the network, and supplier assessments. Predictive analytics can be applied to demand forecasting to reduce safety stock, with a link to both logistics capacity and the organization’s production planning. Freight rate analysis enables better procurement decisions, predictive maintenance of any logistics assets, and with rising congestion in cities, also helps identify the optimal route for a delivery to a customer. Any B2B and B2C flows are key in analyzing distributor and consumer buying patterns, again enabling a supply chain network that sets customer experience and service levels as top priorities, and does so cost-effectively and sustainably.
In this new analytics environment, there is much food for thought for organizations (raising issues that may still need to be answered with judgment, wisdom and experience): How can analytics be a competitive advantage? What ecosystem is needed for success? Are organizations ready for open sourcing to leverage tools and methods? How will intellectual property guidelines around this evolve? How will organizations engage analytical talent?
The deeper we look within our organizations, the more use cases and data we will find. A customer recently asked me whether we need analytics – just because we have the data, and now the ability to mine it, does it mean we must? It is a relevant question. The key lies in identifying the right use cases: Is the problem statement clearly defined? Is the problem business-critical enough? Do we have enough data? Is this available in the clean format required in order to apply sophisticated analytical tools to it?
As long as we keep asking ourselves this question for every use case we apply analytics to, and approach it with the openness and curiosity of an explorer, we may be surprised at what we find. Let the wanderlust continue!
Published: September 2018