Five insights-based strategies that can help organizations launch DAO programs quickly and establish a data-centric culture
Remember Rip Van Winkle? This legendary character from Washington Irving’s 18th century classic fell into deep slumber—missing out on major changes to his life and the world at large, including the American Revolution. As the data ownership moves towards business, it is not advisable for data and analytics professionals to sleep through this dynamic shift underway.
With data playing an ever important role in decision-making, business users now need larger volumes as well as more control over how they access, analyze, present, and share it. As a strategic response, technology players are providing easier-to-use data visualization, exploration, and preparation tools that require less IT handholding.
More importantly, an increasing number of IT professionals and business users are collaborating to set up Data Analytics Organization (DAO). These DAO centers are responsible for identifying, ingesting, transforming, and preparing internal and external data assets. The sole purpose of such DAOs is to manage the rampant democratized data needs of users that self-serve their analytics processes.
Here are five critical strategies that can ensure a successful DAO:
1: Establish a horizontal COE to deliver on service goals
At the enterprise level, DAOs will have to meet higher expectations pertaining to various data and analytics services. To support this burgeoning demand and ensure adherence to service goals, it is imperative to identify the right capabilities, hire staff with the right skill sets, and prioritize rollouts.
2: Make standard tools, methods, and processes available
Compared to where technologies were a decade ago, the current set of tools that support data engineering, data wrangling, visual analytics, modeling, and so on, are simpler to learn and use. However, with progress comes a new set of challenges. Way back in April 2006, Hadoop 0.1.0 was released. By February 2016, there were more than 900 Big Data tools. This makes selecting the right set of tools quite difficult.
The need of the hour is to set up a Tools and Architecture Board (TAB) comprising of both IT and business representatives. TAB will have a broad mandate to drive identification, evaluation, training, and other adoption related aspects of organization-standard technologies and tools. TAB must also ensure all technology architecture decisions are evaluated and are in congruence with the organizational enterprise strategies, business needs, and integration standards and processes.
3: Preserve business value and user personas at the core
Firstly, it’s important to ask: What insights are you seeking? What values will these insights bring to your organization? And, where can you access the right datasets to generate the right insights? In order to architect, set up, and operationalize DAO, enterprises must first identify business use cases, and prioritize them for execution. To create a sustainable, scalable, and appropriate DAO, it is important to:
Thoroughly understand the many personas of data citizens who will be using the ecosystem
Have a prioritized business use case list that can quickly start generating value for the organization while at the same time encouraging adoption
For this, the implementation process must consider the business motives, expectations, working platforms, technical skills, and areas of interest. Adequate measures must be taken to factor in the industry trends, market competition, technology evolution, and the regulatory environment.
4: Balance democratization with proper security and governance
With data democratization, there rises a need for adopting and implementing the right set of governance and security frameworks. This will ensure organizations’ data assets are adequately secured, and in line with regulatory and compliance requirements.
5: Simplify and modernize all aspects – from data sourcing to consumption
End-to-end data management will comprise data identification, discovery, registration, exploration, ingestion, curation, data quality management, data preparation, and data consumption. These streamlined processes must make data requesting, wrangling, and consumption capabilities transparent, and easy to use. Automated processes must be identified and implemented to assist data scientists in registering and downloading data assets. This will allow them to move data into their sandbox environments to perform exploratory analytics.
As data based disruption accelerates, there is a sense of urgency among executives to figure out the strategy and approach towards going digital. Setting up a Data and Analytics Organization is core to going digital. Based on the vision, strategy, and goals of your enterprise, you must identify few Key Performance Indicators (KPIs) for your Data and Analytics Organization. This will enable your DAO to think about the low hanging fruits, demonstrate quick wins, and use these KPIs as springboard to articulate and amplify their success. What can be a better way for a DAO to demonstrate success than a data backed track record of success?