What Is a Data Strategy and Why You Need One

A 2020 Deloitte survey found that 67% of companies are using machine learning and 97% are using or planning to use it in the next year.   When you stop to think about that it is a remarkable statistic – particularly the latter part.  It seems a consensus has been reached and the use of machine learning and advanced analytics techniques are no longer optional – they are mandatory for companies that want to remain competitive or, better, to leapfrog their competition.  Companies are increasingly looking to data and machine learning to increase sales, improve customer experience, reduce costs, increase efficiencies, accelerate product innovation and define new business models.

Yet, despite the importance of data and machine learning many companies do not have a data strategy.  We know that very little of the data that is produced by a company is ever analyzed.  Most companies are not aware of their data assets and they are not clear how to leverage their data to increase revenues and gain competitive advantage.  Due to the lack of a data strategy, they wind up taking a haphazard, project-based approach to improving their company’s use of data.  So what is a data strategy?

A good data strategy will serve to guide an organization and unify their activities to optimize their return on investment (ROI) related to their data initiatives.

The Data Strategy is implemented as a program, not as a project, and recognizes and codifies the importance of data to the organization’s strategic goals.

The Data Strategy is a living and breathing document that serves as a reference point and a framework within which data-related priorities are set and decisions are made.

It is important that the Data Strategy not be just a theoretical exercise full of high-minded aspirations and well-intentioned principles that never get acted upon.

A well formed Data Strategy must include the following three components at minimum:

1. Strategic Objectives and Guiding Principles
2. High Priority Data Use Cases
3. Implementation Roadmap

I will address each of these sections below:

1. Strategic Objectives and Guiding Principles – in this section, the data strategy should describe what needs to be achieved and what the guiding principles are for decision making and execution.  This section should describe how data relates to the strategic business goals of the organization.  It is important to cover all aspects of data and how data will support and enable the strategic goals of the business.  At minimum, this section should include the following:
   1. Customer-related improvements
   2. Innovation and product-related objectives
   3. Internal process-related efficiencies
   4. Data Sources (internal and external)
   5. Data Governance
   6. Data Quality
   7. Master Data
   8. Data Monetization (if applicable)
   9. Analytics Access and Delivery Patterns for Different Audiences
   10. Conceptual Data Architecture (vendor/technology agnostic)
   11. Organization Structure and Roles/Responsibilities
2. High Priority Data Use Cases – the preceding section of the Data Strategy described the strategic objectives and guiding principles for a number of important data-related areas.  However, it did not describe any specific data use cases that will have a significant impact on the achievement of strategic business goals.  In this section we recommend that you get specific and describe, in some detail, about 4-5 data use cases that align to the organization’s strategic business goals.  More specifically, for each data use case we recommend identifying the following:
   1. How it relates to the strategic goals of the business
   2. What is the overall objective
   3. How will success be measured for this data use case
   4. What data will be required and what are the data sources
   5. What analytics and key performance indicators (KPI’s) are part of the data use case
   6. What skill sets and capabilities are required to deliver on the use case
   7. What technology challenges exist
   8. What data governance, security, privacy challenges exist
   9. Who is the data use case owner/sponsor
3. Implementation Roadmap – as we mentioned it is important that the data strategy not be just a theoretical document that sits on the shelf and is not acted upon.  As part of the data strategy, we recommend turning the strategic objectives, guiding principles and high priority data use cases into a program roadmap of specific implementation projects.  In order to create this implementation roadmap, there will need to be decisions made about prioritization, sequencing, resource allocation and organizational impact.  Typically an implementation roadmap will cover about 3 – 4 years.

You may have noticed that we have left out of the data strategy making decisions on specific technical platforms or cloud providers.  We have found that it is more beneficial to consider how data can and should support and enable the strategic goals of the organization without constraining the analysis to specific technical platforms or vendors.

After completing the data strategy we recommend developing a technology and platform roadmap to support the objectives and goals of the data strategy.  This technology and platform roadmap does not specify specific technologies for individual projects but rather outlines a set of preferred technologies based on a number of criteria including but not limited to interoperability, cost, ongoing maintenance, in house skill sets, technical capabilities.

If you would like to discuss plans for development of your data strategy please do not hesitate to contact the Apps Associates team.

About the Author

Myles is the Vice President of Data, Analytics and AI at Apps Associates. He has 30 years of experience designing, building and deploying large scale data warehousing and business intelligence systems. Myles is a thought leader in the area of data strategy, modern data architecture and advanced analytics and frequently speaks at industry conferences. Before getting back into consulting about 9 years ago, he held senior executive positions in the Financial Services and Medical Device industries. Myles has significant international experience having served in Director roles in Tokyo, London and Hong Kong. He specializes in helping companies extract value from their data and transform themselves into data-driven organizations.