The data-driven organization: where to start?

Today many companies thinking about using Data Analytics, Machine Learning, and Artificial Intelligence. These definitions became buzzwords for most organizations, but many don’t pay attention to the fundamental driver for all of them — the data. Without it, you can’t create a good ML model to improve your business, and you cannot get any statistical insight.

Working on several data projects, I faced different issues:

• the understanding and expectations from data usage and governance were different between departments;
• data activities did chaotically, without any strategy;
• the data quality was significant only for the team that works with data.

To avoid these issues and earn the most of the data, use it as a powerful company asset, Businesses should transform together with the technical department. It is not enough to store the data — all the processes related to data should be clear from the business perspective. It would be best if you started with business people and business problems.

Source: https://i2.wp.com/adniasolutions.com/wp-content/uploads/2020/09/shutterstock_1483995179.jpg?w=1000&ssl=1

There are few crucial steps to begin with:

1. Review the current state
2. Identify business drivers
3. Create a Data Governance Strategy
4. Create a transition plan to become a data-centric organization

For organizations that don’t pay enough attention to these steps, the goal to become data-driven and use advanced analytics is neither hard to achieve nor reachable.

Let’s take a look on a high level at these steps.

Review the current state

Before creating a new organization or improving the current one, it’s essential to understand the current state related to data and the existing operating model. The crucial questions are:

• How are decisions made?
• Are they data-driven?
• Who manages the data?
• Are there defined owners of the data?
• Are KPIs for the data in place?
• Are there any Data Strategy?
• If yes, is it in line with business strategy?

Answers to that questions will help identify roles, responsibilities and decision-making processes for data management.

Business drivers

Data is an important company asset. But it is hard to measure the value of the data. You can’t say “I have 12 GB of data; the profit of storing it will be 100$/month”. Usually, data is deeply integrated into different business processes, so it’s hard to identify exact value. But we can identify the business goals that use the data and find out what data is needed.

Identifying the drivers will help you decide on the complexity of the needed Data Solution and store only relevant data. It will be the core for estimating the gained value and setting up data KPIs in the future.

Data Governance Strategy

Data Governance aims to ensure that data management is done correctly, according to company policies and best practices.

The scope of Data Governance Strategy is individual to each organization, but most of them include:

• policies on how to manage data and metadata;
• compliance requirements;
• data quality and data architecture principles;
• issues management;
• evaluation of data as a business asset.

Different company levels should support understanding and implementation of Data Governance Strategy — the organizational culture must learn to value the data and related activities. It’s a continuous process that requires organizational commitment to change.

Organizational change

Improving data management requires changing how people work together, understanding the data’s organizational role, and using data to obtain business value.

For example, suppose your IT department consists of both of data producer and a data consumer team. In that case, both sides should enforce data quality. They should work together with the business stakeholders to identify and implement quality requirements.

The change by itself might be complex. People already worked in some way, and it should be clear why they need to change, the steps, and the final goal. For those who are interested in change management, I would recommend reading about Bridges Transition Model.

Conclusion

Before using the data, the organization should define main processes, requirements and KPIs related to data management. In most cases, it requires an organizational change, where the business should transform together with the IT department. Strong integration of Data Strategy into the organization Business Strategy will help get the most value of the data.