Solving Problems, Not Symptoms: Acknowledging Data’s New Role

Lately, I’ve noticed a recurring theme in data discussions: “Solve problems, not symptoms.” While this sentiment resonates strongly, it’s easier said than done, especially in the context of data. The reality is that solving data problems often means invoking changes upstream—before the data team even receives the data. And this reveals a deeper organisational truth: many businesses still treat data as a supporting actor rather than a central pillar.

In my previous article, (From Foundation to Finish: Building Successful Data Transformations That Last) I explored why data transformations often fail. As a natural follow-up, let’s examine what’s required to truly “solve problems, not symptoms” in data. Here’s the understanding businesses need to embrace: data is no longer an optional resource—it is intrinsic to every part of your organisation.

The Evolution of Data in Business

To set the stage, let’s start with some context:

1. The data revolution is unlike other disciplines. Data’s rise to prominence began in the early 2000s, but unlike disciplines such as engineering or finance, its foundations don’t lie in academia. While machine learning (ML) and deep learning (DL) stem from decades of academic research, data as a business function has been shaped by rapid technological evolution rather than established theory. This makes it harder for organisations to instinctively understand its value or integrate it effectively.
2. Data is not just a reporting tool anymore. Historically, companies used data primarily for reporting and descriptive analytics. However, becoming a truly data-oriented organisation (more on that term below) is not about sprinkling machine learning or AI on top of your existing operations. It’s about embedding data into your processes, culture, and decision-making from the ground up.
3. Let’s talk about “data-driven.” Personally, I’m not a fan of the term “data-driven.” Unless your entire product is data, you’re not “data-driven”; you’re data-oriented. The difference? Data-driven implies that data dictates decisions without nuance. Being data-oriented means incorporating data thoughtfully and strategically across your organisation.

Why Solving Data Problems Requires Organisational Ownership

Here’s the crux: to solve data problems, organisations need to address issues at their root—often far beyond the purview of the data team. For example, ensuring the success of machine learning, deep learning, or generative AI projects isn’t just about having a great model. It’s about having high-quality data from every department, where quality encompasses:

• Accuracy: The data reflects reality.
• Completeness: No critical gaps exist.
• Consistency: No contradictions across datasets.
• Timeliness: Data is available when needed.
• Uniqueness: No unnecessary duplicates.
• Validity: The data aligns with business rules and standards.

These qualities cannot be ensured by the data team alone. Here’s the key: every department must take ownership of their data. Marketing, finance, HR, operations—each has a role to play in ensuring their data is clean, accurate, and accessible. This is where a robust data governance policy becomes crucial and where the value of data stewards is truly realised.

How to Get There: Building a Data-Oriented Organisation

If your organisation truly wants to “solve problems, not symptoms,” it’s time to redefine your relationship with data:

1. Elevate data as a strategic asset. Stop thinking of data as a back-office tool. Every department should understand how their data impacts not just their team but the entire business.
2. Foster cross-functional accountability. Data is not the sole responsibility of the data team. Establish clear ownership for data quality and availability within each department.
3. Prioritise data literacy. Empower employees across all levels to understand, interpret, and utilise data effectively. This bridges the gap between technical teams and business functions.
4. Invest in data governance and collaboration tools. Ensure processes and technologies are in place to maintain high-quality data and streamline communication between departments.

Final Thoughts

“Solving problems, not symptoms” in data requires a fundamental shift—one where data transitions from being a reactive tool to a proactive foundation embedded in every part of the organisation. Without cultural integration, even the most advanced technological enhancements risk becoming technical debt rather than delivering real business growth. By fostering cross-functional accountability, prioritising data quality, and embedding data into your organisation’s core processes, you create an environment that identifies and addresses root problems rather than merely treating symptoms.

Has your organisation truly embraced the role of data, or is it just enhancing it's technical debt?