Thousands of business decisions today are based on analytics. Data is fast becoming the common medium of commercial conversation, where dashboards replace PowerPoint decks as the standard visual reference point for meetings.
It’s also making every organization more data centric. That's why data self-service is so important: It takes analysis out of the hands of specialists and democratizes insights across the business.
Everyone can benefit, but a lot must happen to make self-service work:
- The data must have well-designed semantic layers
- The delivery mechanism that sits behind the BI dashboard must have its frequency, volume, and velocity established
- The data must be trustworthy, with its quality, compliance, and lineage firmly established
But how do you get to this end state? Whether you’re migrating data to the cloud or modernizing what’s already there, there is almost always a gap that needs to be filled. To be successful, it’s vital to get your hands on data sets as early as possible to assess issues like complexity, governance, and the use cases that will determine whether information is valuable.
Mapping affinities to understand value
One of the most difficult issues is deciding what data stays, what data goes, and prioritizing the chosen data. This phase is less about technology and more about methodology.
To help with this process, we suggest affinity mapping, which discerns the value associated with the different use cases and data sets that the business will apply. Affinity mapping also helps identify which target data sources to use first and then matricizes them to understand how one data source can unlock “X” number of use cases.
Affinity mapping can also help secure buy in from the broader business for the migration or modernization initiative. You'll be able to see which business units are hitting which data sources most frequently and predict the benefits each one can expect to derive.
Overcoming the fear factor
Choosing which information to keep—and then throwing it into a shared pool for analysis—can bring a level of anxiety about diminished roles and loss of resources. There’s also a possibility that taking significant costs out of a business unit or division might lead to a budget reduction. However, with modern analytics, opportunities can be identified to pre-empt a budget cut and turn it into reallocation for a new initiative. For example, an initial goal for a business case can go from “We’ll spend one hundred thousand dollars, and we'll save five hundred thousand dollars,” to “And we'll then reinvest that five hundred thousand dollars to help solve another issue.”
Removing data from silos and moving it into a broader, more centralized space might be daunting at first. But once people see what can be done with common access to analytics from a rich pool of company-wide information, they will quickly see the value. The key is to establish trust in the information that a business user hasn’t seen before or had any control over creating.
Governance as a business opportunity
If you're thinking about data modernization or migration, it's also the ideal time to address data governance. The right governance framework will enable the desired self-service capability to establish when the modernization/migration is complete. Plus, it's much easier to establish a framework at the outset rather than play catch-up later.
Establishing data governance should be managed in three phases:
1. Establish key roles and forums. There has to be a place where stakeholders can discuss the challenges and opportunities around making data governance happen. It's vital that everyone knows, for example, who the data stewards are, who the company domain experts are, and so on.
2. Establish standards for data lineage. The goal of governance is establishing trust, which will later become the basis for collaboration. Knowing where the data has come from is vital. From a tooling perspective, that's not always easy to achieve, since data points can exist in billions of different combinations.
Informatica’s AI-powered data lineage solution ensures you capture all the relevant metadata about all your data—regardless of the source. This ultimately gives you a detailed, end-to-end view of data lineage across the cloud estate.
3. Establish standards of data quality. Once affinity mapping is completed, data lineage is established, and the business value and use cases attached to your data sets are understood, it’s much easier to eliminate the “garbage.” You can remove data that could confuse analytics or reduce the strategic value of insights.
When done right, cloud migration and modernization can accelerate innovation. By enabling self-service analytics, both can offer users across the business more power to identify transformative business opportunities on their own.
Whether interested in migrating or modernizing your data, Deloitte, AWS, and Informatica are ready to help you take your data to the next level. Learn more now.