< Articles & News

Analytics Of Things Capability Model

August 1, 2016   |   9:02 PM
Big Data Analytics
Teradata Articles

by Cheryl Wiebe,Teradata

How do you know whether you have the right capabilities to handle Analytics of Things to support your company’s IoT strategy?

chart-and-ladder

In the first two posts on this topic, I posed and answered the questions “ what is Analytics of Things” and “ Are you Ready for Analytics of Things.” I previewed the continuum of capability, which can be depicted in levels with signposts along the way (see below).

capability-model

As promised, I’m going to elaborate on what are the capabilities, and what must organizations achieve at each stage to progress to the next. Originally I was going to give all this in one monumentally long post, but I decided to set out for you the first three stages first, below, and next month wrap up the final stages. For many organizations, these first three will be relevant to where they are—getting started.

The Novice: Align strategy to understand how to progress

noviceThe organization is just getting started with analytics and ’things’. Although they have implemented some rudimentary sensors in their operations, they have not built processes for collecting and systematically managing the data. Their focus is on the tactical infrastructure: the sensors, the edge nodes, and operational solutions. There is no clear investment strategy for IoT in their corporate or departmental business plan. Data, technology, and strategy are all disconnected yet a few silos are emerging. To get to the next level of capability they must align their IoT strategy to their business strategy, formulate a roadmap, and formally justify investments in their business unit. A major step forward is to provide business users sensor data accessible by business intelligence tools. Start the learning process.

 

The Apprentice: Experiment and build analytics capability

apprenticeAt this stage the organization is operating edge nodes and some rudimentary operational efficiencies emerge. The business unit’s focus is on managing the “things”, typically at the department level. Investments are made at the business unit or departmental level but are isolated from the overall company strategy for IoT. Insofar as data is being managed, it’s in small, isolated pockets, data puddles, and data marts. Departmental data marts emerge and BI tools are in use. To grow its capability this organization needs to collect and experiment on sensor data, run experiments on it, build capability in time series data (a whole new class of analytics calling for specialized talents and techniques). It needs to begin to build predictive models and add features and predictors as more and more data is integrated into the models. It needs to build a plan for integrating the sensor data with back office systems, and understand what use cases require what level of responsiveness. These experiments and use case analysis will inform future data architecture by identifying what data and analytics are needed and where they are needed in the IoT ecosystem.

practitionerThe Practitioner: mastering internal operations leads to new digital opportunities

The Practitioner integrates data across from multiple business units and shares the data according to business use/needs. Business units begin to have cross-functional visibility due to organization and assembly of data into a unified data/analytics ecosystem. Organizations with a pre-existing unified analytics ecosystem only need to bring the new sensor data into this analytic ecosystem. Either way, they must build skills appropriate to time-series intensive, sometimes messy, and frequently huge sensor data. They must also build processes that support the cooperation and integration of Operations Technology (OT – in plants, operations, and out at the edge) with Information Technology (IT – traditional business systems in data centers, and front/back offices). Benefits at this stage accrue rapidly as the organization can improve its internal operations and lower aggregate costs and sub-optimal processes. Although the Practitioner has taken meaningful steps toward digitalizing its internal business, and taking down barriers between OT and IT, to transcend this stage, it must begin to prove its capabilities for providing information-based services, such as early warning services, displayed on digital dashboards, or prescriptive “next best action” recommendations, for which customers might pay service fees.

Stages one-two-three set realistic, medium-term goals for companies who have realized the importance IoT has to their organizations’ business strategies. The stages laid out above provide a realistic, albeit high strategic level roadmap for the Analytics of Things strategy to complement those strategies and goals. Assessing and evaluating one’s capability can help set goals, timelines, and project priorities that are suitable at each stage, and ultimately accelerate the progress the organization makes toward its desired capability level. Stay tuned…the next of this series completes the description of the stages, covering Innovator and Game Changer.

thought leaders

Teradata  Thought Leaders
Teradata

View Profile