Start delivering trusted insights — here’s how

Data may be everywhere, but it isn’t free. It takes a lot of work and infrastructure to turn raw data into useful insights. Research suggests that the cost of handling data is only going to increase, by as much as 50% over five years. The same source suggests that part of that cost comes from confusion — users may spend up to 40% of their time searching for data and up to 30% of their time on data cleansing.

The issue here is data trust. If they aren’t sure what data is good or which data sources are reliable, your users are going to spend additional precious time and resources trying to get to the truth. As your company’s data leader, you can save time and money — not to mention frustration — by establishing standards for data trust.

The “5 Ts” of data trust

Your data users need to have confidence that your organization’s data is valid, complete, and sufficiently high-quality to function as foundation for their applications and analytics. At Talend, we use a rubric that we call the “5 Ts of Data Trust” as the basis for any conversation about trusted data.

• Thorough. Is your data clean, complete, and consistent across your systems?
• Transparent. Is your data accessible and understandable?
• Timely. Is your data readily available to the people who need it?
• Traceable. Does your data tell you where it came from and how it has been used?
• Tested. Has your data been rated and certified by other users?

What sorts of problems do you face with your data? For example, one dataset might have a detailed history of where it originated and how it’s being used, but have a few missing fields. Another might be complete, but it’s out of date. A third might be pristine in every way — but trapped in an obscure application that only a handful of engineers can access.

The path to complete data trust is a journey. Data teams will have to make their own assessments about where to start and what threshold various datasets must pass to qualify as trusted data. But it doesn’t take 100% perfection across all dimensions for all tables, records, and fields for data to be trusted. As long as you have shared standards for data trust, your users can be confident using that data.

Getting to trusted data

Have you ever given some reports or analysis to a business leader, only to hear, “This can't be right,” or “Where did you get this?” Trust needs to be the keystone of your data strategy, which requires both technology and building a culture of data trust with the people and leaders inside your company.

Since you manage what you measure, innovative data leaders should think about how their quarterly objectives and goals can better serve a more trusted relationship between the business and its data.

To get to trusted data, you will need to implement and automate processes for auditing, assessing, and cleaning the data. The platform you use should bundle data integration, data integrity, and data governance features — ideally in a single, integrated system. It should leverage the knowledge of line-of-business users to clean data where appropriate and support tools that let data engineers perform complex operations through an easy-to-use graphical interface.

But you will also need to establish new rules and workflows that put data trust at the center. Teams in data and IT are familiar with outlining service-level agreements (SLAs) that describe their obligations to customers within the business. These SLAs commonly govern things such as response time or data availability, or commitments to provide regular training to prepare non-technical users for self-service data use and access. Those things are vital, but they leave out one critical factor: trust.

As you draft new standards for data delivery, include components of data trust in your SLAs.

Your data trust action plan

How do you make sure that you’re accounting for data trust in every step of your processes? It may be easier than you think. You just need to break your action plan into simple, manageable steps and you are on your way.

1. Define a trust metric. You may be familiar with evaluating the health of data itself, but consider how that shows up to the rest of the business. Ask your stakeholders how they think about trusting — or not trusting — the insights they work with. Then define a metric, such as the number of times in a quarter where business users reported a trust-related issue when using existing reporting. That trust metric should be integrated into your regular data reporting.
2. Create a process for measurement. If you don’t have a ticketing system already, consider implementing a process where business users can report problems, such as missing data or broken dashboards. This will let you cloud-source qualitative information about your data quality that you might otherwise miss.
3. Build a baseline. With your process for measurement in place, spend a quarter collecting measurements to form a baseline of how often data trust-related issues impact your business stakeholders. Moving forward, you can use this starting point to measure your progress as well as identify areas where that process might be lagging and require additional help.
4. Set your sights on improvement. With a baseline in hand, you can set goals within your team to improve each quarter. As patterns start to emerge regarding any processes, sources, or tooling that are preventing consistently trustworthy insights, you can put on your product manager hat and tackle them one by one.

Benefits of data trust

Trust is the key to making successful use of your data. When you have a way to ensure data trust, you can get the data you need to design exceptional customer experiences, improve operations, ensure compliance, and drive innovation.

Your data trust infrastructure can automate data quality checks and create reusable processes, saving data engineering time and increasing organizational productivity. With trustworthy data, everyone in the organization can be more confident that they’re making decisions based on a complete, accurate, and timely picture of the real world. Instead of wasting time second-guessing, they can invest more time in acting and making change. When you make decisions using trusted data, you’re likely to make decisions that lead to better outcomes — which then leads to higher revenues. Trusted, accurate data also makes it easier for organizations to meet regulations for data privacy, such as GDPR and CCPA. Ultimately, clean, trusted data lets you respond better to your customers’ needs, which drives improved customer satisfaction.

When you have trusted data available across an organization, you can more easily grow, adapt, and reach your business goals by making informed strategic business decisions with confidence.

Want to know how your data trust measures up? Download our Data Trust Readiness Report to compare your data trust readiness to the benchmark, uncover practical tips and best practices for boosting data integrity, and discover how real companies accelerate data integration and access.