First Line Software is a premier provider of software engineering, software enablement, and digital transformation services. Headquartered in Cambridge, Massachusetts, the global staff of 450 technical experts serve clients across North America, Europe, Asia, and Australia.
The value of data has grown tremendously in recent decades. Businesses today recognize the opportunities that data provides and understand they need to work with it. Yet many organizations continue to use data as a by-product of their activities, rather than a stand-alone business asset. If you want to move to a data-driven business culture, optimize your technology investments, and get the most out of your data—you need to take a holistic approach to all the information that flows through your organization. Forming a data strategy plan would be the first task to achieve those goals. But before we begin, let’s quickly define what Data Strategy means and then do a little business, “reconnaissance” ahead of our 7 steps.
What is a Data Strategy?
A data strategy is a long-term plan that defines how data will be collected, organized, and processed based on business priorities. It describes the technologies, processes, people, and rules referring to an organization’s information assets.
No matter what size of business you have or what industry you work in, you most likely have data you collect and operate with. An effective data strategy makes it possible to increase operational efficiency, optimize processes, make well-grounded decisions, and increase customer satisfaction.
Preparing to Build a Data Strategy
Before you start building a data strategy, you need to do some introspective reconnaissance on your business to define your data maturity, and then select a framework for your data’s organizational structure.
Define Your Data Maturity Stage
The first thing is to determine what stage of data maturity your company operates at today.
You can take advantage of Dell’s popular, “Data Maturity Model” which helps companies determine how data-driven their business is. It offers 4 stages:
- Data-aware: Your company does not have reporting standards. Systems, sources, and databases are not integrated with each other. There is no trust in the data.
- Data Proficient: Trust in the data is still very low, especially in its quality. You may already have a data warehouse, but some elements are still missing.
- Data Savvy: You can make business decisions based on your data, but business leaders and IT specialists continue to work separately. IT works to provide qualitative data only upon request.
- Data-driven: The basic work (in particular, integration of data sources) has been successfully implemented. IT and business leaders work closely together and are on the same page. Now the focus is on scaling the data strategy.
You can use the above stages to be realistic about where your company is today and think of what you want to achieve in the future.
Offense and Defense Data Strategy Frameworks
The second preliminary step that needs to be taken is to determine the focus of your future data strategy and its organizational structure.
Data Defense means focusing on security, access, governance, and accuracy while Data Offense prioritizes gaining bigger amounts of insights enabling decision-making.
If you’re working with highly sensitive data, like in healthcare or financial organizations, you might have to lean on a defense strategy. On the other hand, if you operate in a fast-moving industry like tech and depend upon a quick turnover of data insights, you are likely to go for an offensive approach.
The framework you choose for your business conditions determines the structure you will implement in your Data Strategy.
- A centralized strategy suggests a centralized BI (business intelligence) or reporting team that manages and prepares data and reports for the whole company. This is a slow way to do things and it suits small companies better than big ones, however, you have tighter control over every piece supporting the defense of your data.
- A distributed model gives more opportunities to departments, while BI is responsible for the platforms and setting guardrails. Teams can move faster and do work in a way it suits them. This model is the best for large companies with an offensive approach.
7 Steps to Build Robust Data Strategy
Now that you know what level of data maturity your organization falls into, you’re ready to start building a Data Strategy for your business. Consider taking all (or most) of these next 7 steps to achieve a well-thought-out data management and analysis plan.
1. Define Your Goals with Data
Understanding what your business is trying to accomplish is essential in creating an effective data strategy.
To begin, you must identify an executive leader who will drive the project within the company and stakeholders who will represent individual departments and functions in your organization. This helps ensure that you have buy-in from the executive team and members of the different departments.
Determine the strategic goals of the whole company and the goals of specific teams. Think of how they synchronize with each other and map up to the greater company goals. Find out what the department leaders are measuring, what they are trying to improve, what questions they want answered, and what KPIs can help them get those answers.
2. Inspect Your Existing Data
Once we know what questions we need to answer, we can go further to analyze existing data sources, and understand how the data is gathered currently, and where it exists.
For your in-house data, ask yourself these questions:
- Does it give answers to all of your existing business questions?
- Is it accessible to everyone who may need it?
- Does it have the right level of detail to answer questions effectively?
- Is it updated frequently enough?
- Are there any restrictions concerning the data (GDPR, licensing, etc.)?
Most likely you’ll find some areas for improvement. For the data that is not available at the moment, think of opportunities. Where can it be found? What actions need to be taken to get this data and what technologies may be involved?
Map your business questions to the data sources—both existing ones and those you might get in the future. Note all important information regarding the frequency, details, and regulations of the data.
3. Design Your Data Architecture
Data Architecture turns your business requirements into data and system requirements and defines how that data will be managed throughout its lifecycle. It describes the structure of an organization’s logical and physical data assets and data management resources.
There are many approaches and options to build your Data Architecture. Depending on the size of your business, the level of your data maturity, and other specifications, your Data Architecture can become more complex and forked.
A visual way to represent Data Architecture is to create a diagram, showing all the components involved in your data lifecycle and how they are arranged and communicate with each other. This diagram will identify and represent all the source systems, the methods to ingest data, and the landing spots for that data (data marts, data lakes, and data warehouses). It will also include data governance and information regarding security details.
Once you’ve started creating the structure you can also begin to select the right technological tools for each component and process on your architecture.
Here are some helpful questions to ask in this step:
- How will the data be stored? Should a data warehouse be organized on-site or would leveraging a cloud-based solution make more sense? Does your organization have enough skills and technical infrastructure to expand and support the system itself, or do you need to find contractors?
- How would the data be integrated into the central repository? Is there any standard tool to use? Will you include business logic on this layer so the data will be ready to use?
- How will access to data be provided? Who will create the reports – IT or departments? Do you need reports to be interactive? Will they be for internal use only or can be provided to people outside your organization?
- Are there any tools or sources needed to fill the gaps of missing data? Can it be calculated or estimated with the new technologies implemented?
The more detailed you are with requirements and possible needs upfront, it will translate into a more effective plan and final solution that works for your business..
4. Decide How You Will Turn Data Into Insights
Data in the abstract has no value. Value is born when somebody interacts with the data and gets meaning, sees opportunities, makes decisions, and understands the big picture. That’s why it’s so important that the final presentation is clear and easy to use. Think of the best tools you can apply for that. Here are some things to take into consideration:
- Meaningful Visualizations: Data visualization is not a decoration of a report, but a powerful tool that allows you to quickly extract insights from data. It is important that the presentation does not confuse the user.
- Context explained: It is important that the user understands the context of the data being studied, the history of its acquisition, and its analysis.
- Democratization of data: Consider the background and knowledge level of the users who will view the report. Give definitions for terms and metrics, and explain everything that may be unclear to end users.
- Data granularity: The level of detail required can vary greatly depending on who is reviewing the report. Management wants the big picture, while analysts want details.
5. Train People and Set Up Processes
You can pick up the best tools and implement technologies across all your organization, but your Data Strategy will never work without the support of your employees.
In this step, you have to understand how different people in your company are involved in the data life cycle. They can be end users or data providers (sources). Or both at one time. Make sure that everyone has the necessary knowledge and enough skills to use the tools to make the system work.
Do you think you have enough people at the moment? Perhaps with a more considered approach to data, you will need new specialists or existing employees who will need to undergo additional training.
Pay attention to the structure of the company and the interaction of departments. Perhaps in the new conditions, you’ll need to connect some departments (for example, analysts and IT) and make sure they work seamlessly with one another…
Support and encourage people to work with data. Show them the results of what they can accomplish with it and inspire them with achievable goals.
6. Establish Data Governance
Data governance determines the internal standards of the company (data policy) about how data is collected, stored, processed, and deleted. It describes who has access to what data and who has control over which types of data. Data governance also takes into account the external standards of your industry, government, and 3rd party organizations.
Data Governance programs include:
- A set of standards and instructions
- A plan for the implementation of standards and instructions
The key functions of Data Governance programs are:
- The formation of data entry processes in accordance with established standards (correctness of terms, consistency of business processes, compatibility with similar or related data).
- Identification of persons responsible for entering and/or deleting data, and the correctness and security of this information.
- A unified data management system, i.e. the same standards apply across all business processes.
Effective Data Governance programs reduce risks, increase data security and trust, help streamline processes, and make them more transparent.
7. Create a Roadmap
The end result of all this work is the compilation of a data strategy roadmap. Now that you’ve identified everything that will take you from your current state to where you want to go, it’s important to prioritize.
For each step to be taken, determine the feasibility and expected business value. Find actions that are easiest to implement, but at the same time, can bring quick results for the whole business.
- Investment needs and budget.
- Availability of employees and outside help.
- Other projects that may borrow resources (human and monetary).
- A timeline to mark your progress and victories along the way.
Regardless of the data maturity level you’ve reached, you will need to get back to your data strategy from time to time and reassess it.
You may pull in a new product or feature that will gain you new data issues. For example, you start collecting more sensitive data and have to shift towards a data-defensive structure.
Even if nothing happens, you can track the effectiveness of your Data Strategy based on two indicators:
- Is there frustration with how long things are taking?
- Is there a lack of trust in the data?
If you start noticing any of these two, you may have to go through your strategy once again and find places for improvements.
Ready to Create Your Data Strategy?
Building a robust Data Strategy is a complex task requiring experience, knowledge, and skills. At what level of data maturity is your business today? What would you like it to be?
First Line Software has extensive experience with data projects. We operate in different verticals and understand the nuances within and we’d be thrilled to discuss your data strategy needs. Reach out to us today to learn more!