Time to benefit from a data strategy

Content updated in April 2022.

In this article you will learn:

• Why you should create a data strategy
• How to identify a good data strategy
• How to build a data strategy

What is the purpose of a strategy?

A strategy is a plan to achieve a desired goal. A well-crafted strategy answers the question “How?” and describes what needs to be done to achieve the jointly agreed goals and visions.

Every year company management works hard to sharpen their strategy, yet the staff often sees the result only as a compact PowerPoint presentation decorated with catchy images and slogans. The real need for change, and the means to implement that change, often remain a mystery.

Despite this challenge, strategy does have an important role in data-driven business. It ensures a shared understanding of the goals and commitment of top management. It also outlines the first steps and an initial roadmap to move forward, as well as how to monitor and support the execution of the strategy.

"Strategy has an important role in data-driven business."

Why is data strategy so important today?

Data has become an increasingly important asset for companies. People rarely knock on the door of a colleague, customer, or supplier, nor do they often send mail by post. A modern business communicates through systems, applications, and digital services, and it is data that provides answers to questions like what, when, who, how, why, and what if? 

The higher the quality of its prioritized data, the better and more efficiently a company will operate. Time is not wasted searching for information, and decisions are based on facts instead of on guesswork or gut feeling. The company that can generate valuable insights from their data in the fastest, most effective, and most cost-efficient way will win. A winning data strategy focuses on streamlined information (both data and analytics management) and production processes to support effective decision making.

In real life, data-driven operations may still be a distant dream for many companies. In recent years the massive development of data and analytics technology and the increased availability of alternative data sources (e.g., IoT, sensors, geospatial, 5G) has created almost unlimited opportunities in analytics. Unfortunately, these hyped opportunities often remain unrealized and new initiatives are abandoned after the first proofs of concept. One of the typical reasons is that the data is not fit for the intended purpose, and that is often due to lack of basic data management capabilities proactively ensuring that the data gathered meets the necessary quality requirements.

Beyond ensuring data quality, completely different types of expertise are needed for data security, privacy, the ethical use of data, and the insights generated from it. These areas need to be skillfully managed in order to minimize financial and reputational risks.

This is why companies desperately need a data strategy. A good data strategy takes analytics into consideration, including use cases that have the maximum effect on the business. It also takes data management into consideration, outlining how to make data available efficiently for these purposes, and takes security, privacy, and ethical requirements into account by design.

Want to speak the language? Use our data glossary to master data terminology

Does your company already have a data strategy? How good is it?

There is a need to specify different strategies for different purposes – and all of these should be driven by the business vision and strategy to ensure effectiveness. It is critical to clarify and describe how the business vision and strategy are dependent on data, its usage, and management, and to create a separate sub-vision and strategy for data.

Almost all companies have a business strategy guiding their core business, and most companies have an IT strategy and even a digitalization strategy. These define which digital services or automation initiatives support the business in achieving the objectives defined in the business strategy and how they are to be implemented.

Some companies have already developed a data analytics strategy that sets out how data is used and for what purpose, and what kind of data products are monetized. However, only a few companies have a data strategy or data management strategy – a shared view on how data can be best prepared for the benefit of the entire organization, ecosystem, customers, and other stakeholders.

Effective use of data requires that it is accessible, understandable, and fit for purpose. A good data strategy addresses all these aspects in a balanced manner and is based on conscious prioritization of which data content is most valuable, to whom, and why, and how the data is consumed and managed. Common sense tells us that it is impossible to excel in analytics in an agile, scalable, cost-effective, secure, and quick way if the data is not efficiently managed in such a way that it is ready for consumption.

In order for the company to become data driven, the whole organization needs to start thinking of data as a strategic asset and become more data-centric. Increasing the importance of the business meaning of data and starting to make joint decisions on data´s business content, structures, quality, and architecture is a good start. The strategy should insist that the participants of each critical data and analytics value chain are identified and officially mandated, and that they gather for data-centric decision-making on a regular basis – this is called data governance.

Further reading: What is data governance and what if it did not exist?

A data strategy requires support from top management to become reality

Data management and analytics are not rocket science. However, achieving data-centricity changes the company culture and thus requires a systematic increase in data awareness as well as a change in the operating model and processes. This means educating staff, managing transformation, and most importantly gaining highly visible support for data-centricity from top management.

Many organizations would benefit from having an in-depth understanding of data and methodology in their top strategic management. Some such experts have already been appointed, and companies worldwide are increasingly appointing chief data officers (CDOs). This trend began in the financial services sector where, in the aftermath of the 2008 financial crisis, it became mandatory to focus on data issues. Now other industries are following this lead due to their increased focus on analytics (AI/ML), data platforms, and a digitalization agenda.

Regardless of the title of the position, data management should be the responsibility of a dedicated senior executive, not an additional responsibility for a chief technology officer, for example. A fully committed sponsor from top management is required in order to take responsibility for data management and analytics at the enterprise level. By facilitating and creating a good data strategy, the sponsor ensures that colleagues in the company’s management team buy into the idea. The entire management understands that data affects all operations of the company, making it necessary to collaborate and create common practices across business divisions and functions.

Without the support of top management, companies’ “data debt” will continue to increase, resulting in increased costs, agility becoming rigidity, and increased risks. Often, a data-related incentive and measurement system is needed to truly implement change throughout the entire organization.

Further reading: Data is a great acid test for leaders' digital skills - Explore the success story of OP Financial Group and gain insights into building a data-driven culture.

It is good to acknowledge upfront as a risk that it will most likely be very challenging to get management and organizations to support data-related improvements, especially if those improvements benefit functions and business units other than their own. This is a very normal situation and is caused by silos that naturally start growing around groups of people with different objectives and incentives.

Therefore, the most important purpose of a data strategy is to get senior management to “see the light” and commit to the implementation of a common data agenda across functional silos. At best it leads to a roadmap showing a common prioritization of data initiatives to achieve the goals set out in the data vision and strategy. A data strategy is meant to be a living document that is regularly revisited to ensure it is aligned with the company´s other strategic objectives. It contains both fast results making business happy and hungry, but also persistent activities to improve the ability to manage data.

Further reading: OP's Chief Data Officer Sameli Mäenpää explains the role data plays in Finland's largest banking group and why information is the organization's most valuable capital.

How to get started with creating a data strategy

Tip 1: Remember to outline the clear and visible link to your organization´s business vision and strategy, keeping that connection in mind during the whole process of creating and implementing the data strategy.  

Tip 2: Take your organization’s maturity to understand (data literacy) and manage data into account; it is typical to find that these skills are scarce. It is also well known that experienced business-oriented data experts are a scarce resource, which means there is a need to proactively invest in developing the data skills of your own staff – from top management right down to the individual employee level. Luckily data management is not rocket science – your current organization is capable of learning what is required if it is set as a common goal. Creating a data strategy is a good starting point for this process too.

Six steps to creating your data strategy

1. Analyze the current state

Start by measuring the maturity and analyzing the current state of data management. This helps you to form an understanding of the company’s position on the way towards a more data-centric and data-driven operating model and way of working, and where to start. At this point, you should start looking for members for the company’s “data tribe”. They will be the agents for transformation.

2. Clarify the vision, mission, and business benefits

Clarifying the vision and mission means making sure that your data vision and strategy are derived from the business vision and strategy, and discussing what it will be like when data management and analytics competence are at an adequate level. This helps you to identify how your organization can leverage data to improve its business or streamline operations to reduce costs and save time.

3. Define the governance model

The governance model defines the decision-making process, how funding is obtained, and how progress is monitored. Try to identify existing forums that can add data-related decisions to their agenda. However, pay attention that the participants of such forums are the right ones to discuss data matters.

Change does not happen without someone being made accountable for it, so consider appointing a chief data officer or chief data & analytics officer to take on this responsibility.

Use the existing data management frameworks as leverage, such as EDM Council’s DCAM and DAMA’s DMBOK 2.0. Ready-made frameworks can be used to identify the requirements for efficient data use and management, as well as to identify and outline the important areas for the roadmap.

4. Choose technologies based on needs, considering your company’s maturity level

Implement the most essential data management technologies into the roadmap at an early stage, but not without first analyzing what your company needs and is able to take into use. Too often, applications are not used to the fullest either because there wasn’t a real business need for them in the first place or because the business was yet to recognize the need.

There are some applications that will most likely be essential for the implementation of your data strategy: applications for metadata management and cataloging, master data management, and data quality management for basic data management. Depending on the organization, data sharing consisting of, for example, API management, virtualization, and data platforms including data lakes and data warehouses, may be fundamental. At the early stages, tools for analysis and visualization will most likely be required too.

5. Create the roadmap and indicate the initial costs

Create a high-level roadmap that describes the largest building blocks. Make sure it includes both outputs that generate fast, direct business benefits (effectiveness) and a systematic plan to build a foundation for the longer term (efficiency).  Ensure you take advantage of the initiatives, projects, and proofs of concept that have already been carried out or planned. There is usually a lot to be re-used even if, for one reason or another, the work might have served only a small team or may not have even been completed.

6. Describe the risks of not focusing on data and its management

What happens if nothing is done to improve data management? This scenario usually surprises the company’s management if they haven’t thought about business in terms of data before. Financial, operational, and reputational risks are likely to arise if data and its management are not properly addressed and made part of the business’s everyday agenda.

Does your business need a data strategy?

If you need to get your data in order swiftly, please contact us. Our team is ready to roll up its sleeves and get stuck in!

We are recruiting. Give life to your inner Transformer. Join Tietoevry Tech Services.