Tax & Accounting Blog

How Big Data/Data Analytics Will Transform Supply Chains – Part 1

Blog, Global Trade, ONESOURCE December 29, 2017

Your cell phone receives a text message indicating that import declarations for critical parts required in your China factory missed today’s deadline for filing with your customs broker. The manager of logistics receives an alert that port traffic in Los Angeles may delay by 48 hours the receipt of critical raw materials prompting your logistics software for you to approve diverting the cargo to Long Beach. The director of trade compliance is notified that there is a 90% chance that a critical shipment of parts to Sao Paolo may be subject to inspection by Brazilian Customs, delaying receipt of the parts at your factory by up to two days. Your GTM software recommends shipping the parts to another port to speed up the entry.

What do all of three of these scenarios have in common? They each harness the power of applying data analytics and ‘Big Data’ to improve operational efficiency. Data analytics are the qualitative and quantitative techniques used to enhance productivity and business gains while ‘Big Data’ refers to voluminous amounts of structured or unstructured data that organizations can potentially mine and analyze for business gains. The designation ‘Big Data’ was created as a broad term for data sets so large, or complex, that traditional data processing applications are inadequate.

As corporations continue to face pressure to increase profit margins and shorten order to delivery cycles, the application of data analytics will continue to grow. A Gartner, Inc. study published earlier this year projected 2017 sales of $18.3 billion in the business intelligence and analytics market [1], while sales of prescriptive analytics software is estimated to grow from approximately $415M in 2014 to $1.1B in 2019 [2].

Big Data is defined by the three V’s: volume, velocity, and variety. The first, volume, relates to the sheer magnitude of data currently available for analysis. While we normally think of data as text or numbers, data also includes email, tweets, other content generated by social media, images, audio, scans, etc. In fact, data is expanding at a rate that doubles every two years while human and machine generated data is growing 10 times the rate of traditional business data. IT World Canada projected that by 2020, the sheer volume of the world’s digital data would fill a stack of iPad Air tablets that would extend from the earth to the moon.

The second “V”, velocity, refers to the frequency of change in data. Think of how data velocity has accelerated in the past 5-10 years driven by the expansion of the Internet and social media. Real time data is projected to grow ten-fold by 2025. A cousin of real time data is near-time data- transmissions which include a time delay between the occurrence of an event and the publication of that data. If you have ever accessed a website which provides stock prices published on a 5 minute delay, you have accessed near-time data. Streaming, is a term that probably most of us never heard of before the advent of consumer services such as Spotify or Netflix and the widespread adoption of WebEx in the business realm. Driven by the availability of cloud-based solutions, the growth of streaming services will only accelerate as younger generations have embraced the technology to access movies, music, videos, and television. A study recently found that 61% of young adults ages 18-29 primarily watch TV via streaming services.

The final “V”, variety can be further defined as structured. Semi-structured, or unstructured data. Structured data is data that has been organized into a formatted repository, typically a database, so that its elements are accessible for processing and analysis (think of Excel spreadsheets).

For an understanding of semi-structured data, think of CSV (comma separated value) files. They aren’t parts of relational databases but they are organized in a format that can be easily loaded into an analytical tool such as Excel for analysis.

As the name implies, unstructured data is not contained in a database or some other type of data structure. It may consist of text, numbers, dates, video, images, etc. Examples of unstructured data include:

  • Writing-Textual analysis of written works such as books and blogs.
  • Social Media – blog posts, tweets
  • Natural Language – voice
  • Photographs & Video
  • Communications-emails
  • Scanning communications such as emails to detect spam.
  • Science – Looking for patterns in interstellar radio messages in order to discover intelligent life.
  • Health – x-ray images, scans
  • Analysis of x-ray images for signs of disease.
  • Search – A search engine that spiders unstructured web pages in order to understand their content.

Big Data offers large-scale opportunities to organizations, as both structured and unstructured data can be consolidated and analyzed from multiple perspectives. These perspectives reveal insights while guiding companies in revealing not before observed solutions to complex problems. These new insights will guide companies to scale their programs by combining data analytics with other applications, therefore by also embedding intelligence in every process.

[1] Gartner Says Worldwide Business Intelligence and Analytics Market to Reach $18.3 Billion in 2017, Gartner press release, February 17, 2017

[2] Gartner Forecast Snapshot: Prescriptive Analytics, Worldwide, 2016, February 5, 2016