High-quality data is the foundation of reliable decisions, regulatory compliance and business innovation. Yet a data quality program is rarely a one-off initiative; it requires discipline, ownership and a structured method. The one used here is Danette McGilvray’s Ten Steps to Quality Data and Trusted Information, the reference method for executing data quality projects. Her ten steps group into three phases, and they are the backbone of any data quality program that survives past its first year: assessment, awareness, and action.

---

Step 1: Assessment ; Understand the Current State

The journey starts with a clear-eyed view of the current environment. This means:

• Evaluating both business content (accuracy, relevance, completeness) and technical aspects (integration, consistency, security).
• Identifying business-critical processes and defining the data quality requirements they depend on.
• Clarifying ownership by assigning a data owner and at least one data steward for every dataset.

A thorough assessment provides the baseline to compare against business expectations and identify gaps that matter most.

---

Step 2: Awareness: Build the Business Case for the Data Quality Program

Awareness is about making data quality a business priority, not just a technical concern. This involves:

• Establishing a clear business case that links poor data quality to risks (financial loss, compliance issues, missed opportunities). This is step 4, assess business impact, and it is the step most programs skip. It is also the one that turns “IT should fix that” into business commitment. Without it, a data quality program runs on the data team’s energy alone
• Embedding accountability across the lines of defense ; with business, IT, and oversight functions each fulfilling their role.
• Ensuring transparency by documenting business-critical data and KPIs in ways that are comprehensible and accessible to users.

Awareness creates the organizational alignment needed to mobilize resources and sustain commitment. The same logic applies to data governance: start from a costed business problem, not a framework.

Thomas C. Redman “The notion of quality is inherently customer- (or user-) centric. A collection of data is of high quality, in the customer’s eyes, if and only if it meets his, her, or its needs. It is perhaps the simplest way of thinking about quality and it is certainly the most powerful, but it has profound implications. It means that data quality is inherently subjective”

---

Step 3: Action ; Correct, Prevent, Monitor

With priorities defined and commitment secured, organizations can move to execution:

• Curate existing data by remediating issues ; fixing errors at the source whenever possible.
• Prevent future issues by integrating data quality controls early in the data flow, with checks for completeness, correctness, and appropriateness.
• Monitor continuously by reporting on defined controls, ensuring that quality is sustained at the operational, tactical, and corporate levels.

Action turns strategy into practice, embedding data quality into daily operations rather than treating it as a one-off project.

---

Conclusion: Embedding the Data Quality Program into the Business

A successful data quality program rests on a clear principle: the business owns the data, IT ensures the technical foundation, and governance keeps both accountable. By following the steps of assessment, awareness, and action, organizations not only fix today’s issues but build a resilient framework that keeps data quality high, trust strong, and business value flowing.