How to think about data

In IDC’s “FutureScape: Worldwide IT Industry 2017 Predictions” they predict 10 key developments in the tech world in the next 18–36 months, and beyond, that will impact every enterprise's ability to grow and compete. The three that are the most stunning to us are:

  1. “By 2020, 50% of the Global 2000 will see the majority of their business depend on their ability to create digitally-enhanced products, services, and experiences.” We are seeing this everywhere from “customer intimacy” initiatives, to embedding machine learning into products and services. Data and analytics are being used to digitally enhance what we deliver and how it is differentiated in the marketplace.

  2. By 2019, 3rd Platform technologies and services will drive nearly 75% of IT spending—growing at twice the rate of the overall IT market.”  This prediction sounds quite conservative to us. Some independent analysts say that overall, IT spending worldwide has been essentially flat since 2008. All of the growth is coming from new business-led programs around digitally-enhanced products, services and experience—or what IDC describes as the “DX Economy.” (By “3rd Platform” IDC is referring to cloud, big data/analytics, social business, and mobility. This is very similar to the MIT CISR concept of SMACIT: social, mobile, analytics, cloud, and Internet of Things. The details will vary, but some combination of these technologies will form the basis of the platform for digital disruption at most organizations. 
  3. “By 2020, 67% of all enterprise IT infrastructure and software spending will be for cloud-based offerings.” This is a pretty aggressive forecast, but with all the activity in 2016 where businesses, particularly in North America and EMEA have shifted toward cloud applications and analytics it may be accurate.

As companies navigate their own digital transformation efforts, many ask themselves where the data might come from; what to do with it; and where to store this valuable information.

There are multiple areas where your data can be found, such as: CRM, ERP, social media management systems, Google analytics (web analytics) and other places. These all provide specific information that can be leveraged and structured for reports. Identify where and what data sets matter to your goal, and what time frame and other means are needed to obtain them. Remember, you do not have to obtain all of your data—just the right data. Establishing the right data sets in advance will save time, keep the project focused, and the data you pull will be relevant and targeted.

Three fundamental principles of use are present in just about every Digital Transformation:  accessibility, usefulness, and actionability.

  1. Make data consumable. Data science models—and the data they produce—must be easily consumable by the average user who is seeking an answer to a business problem. Having access to easily consumable, real-time insights and visualizations of complex sets of data can unlock new opportunities and revenue streams—and help improve customer relationships and your bottom line.
  2. Make data adaptable. Models should be self-learning and highly automated, so users can get the most from them. The models must learn and evolve, so the data you get are relevant to your users today and in the future. The models and data also need to be accessible through your existing enterprise platforms, so everyone can easily get to them.
  3. Make data transparent. “Black box” solutions that hide their functions from users or that cannot be re-used across your technology ecosystem are no longer acceptable. When users cannot justify or explain why they accepted a solution’s recommendation, they stop using it. Your users must be able to drill down to understand where the data behind the recommendation originated. When users understand the recommendation as well as the reasons for that recommendation, their experience is more meaningful.

As mentioned earlier, a recurring trait of legacy systems is that they rely on data that resides in functional silos. To be successful with a meaningful transformation a thoughtful, comprehensive plan must be developed that includes functional requirements and buy-in across the board. An often overlooked element is the question of governance. Governance of this data becomes necessary in order to have a fluid operational capacity, as well as the ability to run advanced predictive analytics to help foresee any nuances or issues before making decisions.