Written by Chris Sorensen
Highlights
- A central version of the truth is an established understanding of data that is used company-wide.
- Do not overthink your central truth, just get moving and focus on continuous improvement.
- Establish a Center of Excellence to help govern the process of iterating your way to a central version of the truth.
Most organizations agree that a central version of the truth is an essential part of their data journey.
However, many organizations attempt to immediately build the Taj Mahal rather than building incrementally and continuously delivering value. We recommend basing your strategy on the latter.
When developing a central version of the truth, it is best to simply start somewhere and build from there.
Most grandiose plans for a central truth fail because business requirements change frequently enough that what was needed when you started can differ from what is needed at the end.
The goal of an iterative approach is to narrow this gap significantly.
Statistics show that 70 percent of change management programs fail to achieve their goals. Employee resistance and lack of management support are the two main reasons for this.
There is no one-size-fits-all solution to creating a more data-competent organization, but this article aims to discuss how to build a central version of the truth in a way that is likely to gain traction.
And that is to build iteratively from the ground up.
Sections
- How to Iterate Your Way to a Central Version of the Truth
- 5 Necessary Mindset Qualities
- Stages of Growth
- Final Thoughts
How to Iterate Your Way to a Central Version of the Truth
It can be intimidating to plan for establishing a central truth. Instead, we encourage you to start small and build little iterations over time to slowly reach a central version of the truth.
Do not overthink it. Start with the understanding that the path from point A to point B is not a straight line.
There will be some duplication of effort. There will be some conflicts.
Follow the natural ebbs and flow that will eventually lead to a central version of the truth. Just as your business is evolving daily, your central truth will evolve as well.
Here are some tips to keep in mind while establishing a central truth:
- Take building it one step at a time. You cannot create a central version of the truth overnight. Instead of looking at it as one massive undertaking, start with building a solid foundation and continue to grow and evolve.
- Establish a Center of Excellence. A Center of Excellence is the go-to team to resolve differences, foster communication, and can help answer questions when establishing a central truth.
- Follow a design framework. Provide structure for the project. Architect big, deliver small. Using a dimensional model with conformed dimensions will ensure that as pieces are added to your solutions, they will fit together and set you up for consistent data navigation.
- Let people build their own data models. Don’t force people to use rigid frameworks but make them aware of existing data sets that might prevent them from reinventing the wheel and not consistent with the other wheels. A good data governance strategy is fluid and will guide people to the fold when necessary.
5 Necessary Mindset Qualities
Before developing any plans to achieve a central truth at your organization, it is a good idea to do a quick mindset check-in. Without the right mindset, while developing processes, it is easy to lose sight of the main goal. In the case of a central version of the truth, we want to present relevant and reliable information to business users so that they can use it to make better business.
The idea of creating a central truth can quickly turn into a piece of art to be admired behind glass, but it’s really a tool for people to use, improve, and evolve.
Here are five mindset qualities to consider:
Iterative mindset: Business and processes change quickly, so it’s only logical that your central version of the truth can adapt to strategy changes.
Continuous improvement: Each iteration should result in a slightly better version of the truth. Think of this as a perpetual work in progress rather than a single-end goal.
Patience: This is huge! Developing a central version of the truth takes time, so patience through every iteration is necessary.
Communication: The hardest part of digital transformation is getting everyone on the same page. Management and departments must communicate with each other to garner success.
Executive buy-in: None of these things can happen without a high-level buy-in to the goal of becoming more data-driven and building a data culture.
Stages of Growth
4 key business processes will help guide you on your journey to a central version of the truth:
Center of excellence: A center of excellence houses information needed to upskill people on how to use the environment, collaborate, and how to use self-service tools effectively. The entire organization must commit to using the center of excellence to unlearn Excel habits, unlearned siloed development habits, and work more collaboratively.
Self-service: Establishing self-service enables skilled and knowledgeable users to creatively work with certified data sets. Users can access data when they need it, and rest assured that it is up-to-date.
Data Governance: Data governance provides a map of the existing data foundation for sponsors, owners, and SMEs. It establishes a corporate library and definitions to ensure that everyone within the organization is on the same page regarding reports and data sets.
Dimensional modeling: The data model will be the center of your Power BI universe, and someone needs to own/manage the data foundations.
So how can Power BI help here? The overarching goal for any Power BI environment should be fewer data sets and fewer reports. Sounds simple enough, right?
Excel spreadsheets tended to breed like rabbits as work is continuously repeated and duplicated for a variety of reasons, but most often due to limitations within Excel when it comes to providing mechanisms for consistent analytics across an organization.
A combination of people being aware of the features in Power BI along with the desire to collectively achieve a central version of the truth is needed. If the organization does not care, Power BI will be able to do very little to help.
Here are a few key features within Power BI that can help achieve a central truth:
Certify data sets: Certified data sets ensure that teams across the organization are accessing the same data and building reports based on fewer data sets. Datasets should progress from you and a few users to reach more and more people in the organization until there is an eventual endorsement of the data.
Consolidate and share reports: Sharing reports can break down silos within the organization. It also cuts back on individuals and departments creating new reports and data sets when they already exist somewhere else within the organization.
Establish a Center of Excellence: CoE’s can reduce noise to clean up data that doesn’t agree and fix data quality issues. They promote powerful features within Power BI that may not be known to most users.
Additional features include:
- Data discovery via apps
- Shared data sets
- Row and/or object-level security
- Endorsement
- Apps and shared workspaces
Final Thoughts
A central version of the truth can streamline decision-making by establishing an understanding of data that is used company-wide.
A central truth is best reached over time, through small iterations that happen as the organization grows and changes.
Do not let the term data warehouse scare you. Instead, start small and iterate. Start with one business process and several dimensions first.
Processes like establishing a Center of Excellence, following a framework, and allowing people to build their own data models can help to eventually develop a central truth.
