Data Management

© Copyright 2026: The Intelligent Business Company Ltd, publisher of Housing Technology

The best ways to improve data management activities

There are two sides to the challenge: preventing data issues from arising and improving the quality of existing data.

Prevention often requires more rigorous business process review before implementing new systems. Too often, procurement exercises are used as a substitute for this, when in reality they should follow a clear understanding of how the organisation needs to operate.

When it comes to improving existing data, the most effective approach is usually to focus on the datasets that directly support service delivery, such as repairs, compliance, and complaints, rather than attempting to address everything at once.

Clarity over data ownership is also critical. Many organisations still operate with multiple versions of the truth, driven by unclear system responsibilities.

Importantly, many data issues originate in underlying processes. Improving how information is captured and flows across the organisation often delivers greater value than introducing new tools. Supplier choice plays a role here, particularly in terms of how well systems integrate and share data.

 Examples of poor data quality vs good data quality

Beyond obvious issues such as missing or conflicting records, poor data quality in housing is often more subtle. Assets are not static, their condition changes over time, and not always at a consistent rate. As a result, outdated data can sometimes be more problematic than no data at all, as it creates a false sense of confidence.

The age of data is therefore as important as its origin. While full stock condition surveys are often targeted over a five-year cycle, this remains an aspiration rather than a consistently achieved standard across the sector.

Even professionally collected data is not immune to quality issues if surveys are poorly designed, executed, or disconnected from operational needs. Validation is essential to ensure reliability.

Good data, by contrast, is consistent, current, and supported by clear evidence, such as photographs, documents, or certification where appropriate. Poor data is often not incomplete, but misleading: for example, repairs marked as complete when issues persist, or inconsistent categorisation that undermines reporting and decision-making.

Challenge processes / Choice of supplier

There is a tendency to digitise existing processes without questioning whether they are working in the first place, which often just embeds inefficiencies. Many organisations also underestimate how much their systems contribute to data silos, even when those systems are relatively modern. In practice, the real difference between suppliers is not just functionality, but how flexible and open they and their systems are, and how well they support integration and changing integration needs. Without that, it becomes difficult to build a joined-up view of data across the organisation.

Adoption of external data standards

External standards such as HACT and the Open Data Exchange provide a useful reference point, particularly for organisations that have not yet established a clear internal data structure.

However, adoption across the sector remains relatively limited. Many providers continue to work with data models shaped by legacy systems and historical practices.

While variation across organisations is not always significant, greater consistency in core terminology, for example around assets and components, would help reduce misunderstanding both within and between organisations.

Considering data lakes?

Data platforms, including data lakes, can offer value, but typically only when the underlying data is already well understood and reasonably well managed.

Without that foundation, there is a risk of simply centralising poor-quality data and increasing complexity. In practice, many organisations gain more immediate benefit from developing focused reporting layers around key datasets before moving towards broader platform solutions.

Timing is important. These approaches tend to deliver the greatest value when data ownership is clear, quality is improving, and continuous data management is embedded in day-to-day operations.

There is also a broader point around mindset. The drive to implement new solutions can sometimes detract from the ongoing improvement of existing systems. Today’s platforms can quickly become tomorrow’s legacy if they are not continuously adapted to meet changing needs, and this applies equally to data platforms and their associated tools.

Improving data quality

We see tenders of varying quality across the sector. Many are simply a wish list compiled in the procurement exercise.  Some are the result of rigorous business process reviews by experienced data and business system specialists.  Often IT system changes become a surrogate for business process reviews to drive change.  The number of legacy systems and frequency of system changes testifies to this.

Improving data quality starts with clarity on what data is actually needed to support operations and decision-making, rather than attempting to collect everything.

Across the sector, there are still notable gaps in core asset and property data, particularly where information has not been physically validated for some time. Introducing consistent identifiers, such as UPRNs, can significantly improve the ability to link data across systems.

Equally important is how data is captured at source. Frontline staff and contractors are often responsible for collecting critical information, so ensuring that processes are simple, consistent, and embedded into everyday workflows can have a substantial impact.

More broadly, organisations that achieve the greatest improvements tend to combine strong process design with informed system selection, rather than relying on technology change alone.

Common barriers to data integration

The most common barriers are typically linked to the flexibility of existing systems and suppliers, particularly in environments shaped by mergers, where multiple systems need to be aligned.

Closed or restrictive supplier ecosystems can limit integration capabilities, while unclear data ownership internally can make it difficult to coordinate a coherent approach.

Integration requirements also evolve over time. Systems and suppliers need to be able to adapt to changing data flows and business needs. This flexibility can sometimes sit in tension with the appeal of large, fully integrated platforms, which may take significant time to implement and adapt.

How to improve data governance

Data governance is often approached as a policy exercise, but it becomes effective when supported by clear ownership and accountability across the organisation.

Stronger organisations embed data quality into routine performance management, rather than treating it as a separate or periodic activity. Governance is also more impactful when it is directly linked to key risks, such as compliance, tenant safety, and service delivery, rather than being viewed as an administrative requirement.

End user engagement in data management

Where engagement is low, it is important to understand the underlying causes. Resistance to data-related processes is often rooted in usability, perceived value, competing priorities or imposed change, all of which need to be addressed through effective change management.

Engagement improves when data capture is straightforward and embedded into everyday tasks, rather than seen as an additional burden. Frontline staff and contractors are central to this, as they are often responsible for collecting the most critical data.

Crucially, people are more likely to take ownership when they can see how the data they capture is used, not just for reporting, but to improve services, reduce risk, and support better outcomes in their day-to-day roles.

Click Here to read the latest edition of Housing Technology