Government must address shortcomings to make data strategy work

No uniform approach 

Additionally, the report noted that there is no uniform approach to data across government, with varying levels of maturity displayed across departments – DCMS is the only part of central government where data is referred to as a strategic asset.

“Despite years of effort and many well-documented failures, government has lacked clear and sustained strategic leadership on data. This has led to departments’ under-prioritising their own efforts to manage and improve data,” the report said.

Funding is also an issue that can hamper data projects, with examples of initiatives driven by new policies rather than supporting longer-term, more strategic goals. These include automation of data feeds such as the Department for Work and Pensions using HM Revenue and Customs' real-time information to support benefit payments. 

On the infrastructure pillar, the NAO report noted that legacy systems is a key issue, which makes data extraction and sharing of data – even within the same department – a very complex process and  the combination of different data sources difficult and costly. Data quality in government is often inadequate, with manual effort required to extract relevant information.  

A lack of data standards is another issue found across government, which has led to inconsistent ways of recording the same data, making it difficult for the government to make more use of data in areas such as cross-sector analysis to help understand economic challenges or systemic problems.

Illustrating the scale of the issue, the NAO found more than 20 ways of identifying individuals and businesses across 10 departments and agencies, with no single standard used for recording data such as name, address and date of birth.

Cross-departmental needs

The difficulties in sharing data across government are compounded by a lack of understanding of cross-departmental needs, the report said, adding that obtaining such insight is essential prior to commissioning data projects.  

Commenting on the report, chair of the Public Accounts Committee Meg Hillier said, “Government has been attempting to ensure that departments have access to good quality data for 20 years now, and the NAO’s report pins its failure to do so on strategy and leadership shortcomings.

“Poor quality data is a perennial concern of my committee. Time and time again we see government departments failing to ensure they have the data they should; data which is used to make decisions about how taxpayers’ money is spent, and to understand how effectively it is used,” said Hillier.

The topic of trust around how citizen data is used is a challenge that has also been highlighted in the report. More citizen awareness around the need to keep data secure means departments are more hesitant to look for legal ways to explore the potential of data, despite legislation and DCMS support to departments on how to use the Digital Economy Act to support public services.

Recommendations outlined in the report include using the upcoming data strategy to address the factors hampering improved use of data, including public trust. This exercise should also include a plan to articulate how the government intends to use data and how that approach can provide benefits.

Cross-government accountability, with governance and funding to underpin the delivery of the data strategy, is another point raised in the guidance, with joint working and cross-departmental groups working under a clearly defined plan to avoid confusion around responsibilities.

Rules, standards and common ways to collect, store, record and manage data across departments should also be in place, with regular reviews. An assessment of which datasets are most critical to government should also take place, as well as how they can be improved with process enhancement and automation.

Recognising that departments might differ in terms of their data maturity, the NAO advised organisations to introduce data governance to get a better understanding of their specific issues around underlying data and how data improvement can help them.

The report also recommends departments set out data requirements in business cases, taking into consideration factors such as the current state of information, with thought explicitly given to ethics and safe use of data.

Guidance should be provided for frontline staff on how to use data to help the workforce incorporate government data policy and principles, particularly when it comes to standardisation, data ethics and quality, into their daily working processes.