Data transformation framework: data personas report

5 Workshops analysis

Analysis approach

Two 2-hour workshop sessions were conducted remotely on Wednesday 12^th May 2021. There were 8 participants in each session. The Miro online platform was utilised to capture participant comments and feedback, alongside detailed notes of the discussion by the workshop facilitators. Each participant was given their own colour to capture comments so that they could be traced back to the specific participant to enable persona validation.

Key discussion themes

There were several discussion themes that were highlighted by many of the participants as barriers to their own roles or potential areas for improvement. Although these are all currently well known within the public sector, they are worth documenting as potential areas of initial focus within the Data Transformation Framework. The areas of focus and challenge highlighted by participants also enabled personas to be validated and differentiated.

Metadata and standards

A comprehensive view of well documented available data was highlighted by many participants at all levels of current data maturity as a key driver behind high quality data use and reuse. The lack of good metadata is seen not only to create silos of knowledge and use, but also created a barrier to use, particularly around the appropriateness of use. It was also commented that legacy dataset design often also made some datasets particularly difficult to be used by those not familiar with them.

Data-driven decision-making

The use of data to drive decision-making was seen to be an area with significant scope for improvement. There were two aspects to this: the use of data within existing decision-making processes; and the purpose of specific data collection activities. It was generally agreed that a lot of data is captured, but not all of it is useful for the types of questions being asked. Data and information requests would benefit from additional context around the required decision and ensure consistent and complete data collection processes developed to support this. It was agreed that this would drive trust and transparency in the decision-making process.

Business case for data engagement

It was felt that the business case for capitalising on the value of data had not yet been clearly made. A large proportion of public sector employees are still not engaging with data. Factors that fed into this were the clarity of role profiles, prioritising time within existing roles for data-related activities and the need for upfront investment in upskilling, communication and additional roles.

Data literacy and upskilling

Much of the current training on offer was thought to focus on the tools required for manipulating and visualising data. It was felt that there was a significant need to develop general data literacy and communication across the organisation. The survey also highlighted a significant training need around the areas of data governance, data management, data quality and data ethics.

Tools, software and interoperability

To improve interoperability, reuse and sharing it was felt that the current plethora of tools, processes and systems was not helping. There was a clear wish to use a consistent system such as Power BI for reporting to enable organisations to provide consistent and traceable outputs.

At the same time there some frustration about some of the tooling restrictions in place. This was mainly around access to open-source software such as R or python, which can help to improve analytical capability and auditability for professional analytical teams.

Data sharing and communication

Improved data sharing was highlighted as an area that could break down silos between and within organisation. Visibility of the upstream and downstream data uses could help to bring focus onto the need for improved metadata amongst other things. It was commented that data protection concerns were often used as a barrier to data sharing, and therefore enhanced visibility of data quality issues.

There was a general request for more data to be made open, but the lack of clear guidance around data protection limited the opportunities here.

Data quality, governance and management

Data quality was discussed in detail, with requests for funding for roles for data quality monitoring and remediation to be recognised within any work commissioning process. The importance of ownership and therefore responsibility for data was highlighted repeatedly, but no clear path to achieve this.

Areas not covered

There were a couple of areas that we might have expected issues to be raised, however the lack of discussion on these topics was also insightful. It is acknowledged that the workshops did not cover the full range of expected personas, so other participants may have highlighted these if they had been present.

Data and cyber security

The need for processes around keeping data and systems secure were not mentioned. This is possibly considered as an activity for external teams; however, a growing number of organisations also place responsibility for data and cyber security on individuals.

IT and data engineering

We did not meet any participants who had responsibilities around system management, development or remediation. Data products often require data engineers to automate reporting or set up pipelines for data feeds. Systems often also require complex configurations to ensure they are capturing the right data items and are managing the fully lifecycle of data from capture, storage, definition, access, archiving and deletion.