Developing a Data Strategy
- 1. DEVELOPMENT OF A DATA STRATEGY Turning Chaos into Order: Can It Be Done?
- 2. INTRODUCTION Martha Horler, Data & Compliance Manager at Futureworks A private HE provider based in Manchester with courses in Games, Music Production, and Film. Approx. 470 students, <50 staff, 3 Schools, 9 programmes currently recruiting Responsible for: data returns, systems development, TEF, statistics, enrolment, timetable production, student and curriculum records management, data protection
- 3. QUESTION How many of you work in a data/planning/returns team? What other areas are represented here today?
- 4. WHAT IS A STRATEGY AND HOW IT FITS Strategy: A plan of action designed to achieve a long- term or overall aim. “Strategy is a fancy word for coming up with a long-term plan and putting it into action.” Ellie Pidot “A strategy is necessary because the future is unpredictable.” Robert Waterman “Sound strategy starts with having the right goal.” Michael Porter
- 5. WHAT IS A STRATEGY AND HOW IT FITS Mission: why we exist Values: what we believe in and how we will behave Vision: what we want to be Strategy: what is our game plan Plan: How we will implement and monitor “A vision without a strategy remains an illusion.” Lee Bolman
- 6. WHY HAVE A DATA STRATEGY Issues will arise in any organisation around: -Data quality -Metadata management -Access and data sharing -Ownership -Provenance -Maintainability and Usability -Security and Privacy A data strategy can help you respond consistently to these issues
- 7. PURPOSE OF THE DATA TEAM “The essence of strategy is choosing what not to do.” Michael Porter First task: deciding what the team will do … and what it won’t do
- 8. ACTIVITY List 3 things your team will do with data Then list 3 things your team will not do Which list was easier to create? Which list might be more useful to those outside your team?
- 9. WHAT TO DO & NOT DO What we will do: Build up robust data sources that can be used to produce reliable returns, analysis and meet operational requirements Produce analysis tools that tutors can use to understand the health of their programmes and status of their students Develop processes that automate the flow of data around the organisation and reduce the manual entry of data from other teams to allow them to focus on delivering value to stakeholders
- 10. WHAT TO DO & NOT DO What we will not do: Set thresholds on determining student success, whether that is grade boundaries, attendance monitoring or classification methods Determine how data should be used to support students, this is best done by tutors and student-facing staff Set data retention policies, as this needs to be determined by operational requirements, and approved by the Board of Governors
- 11. GATHERING INFORMATION What data do you collect and store? How is this data currently used? How could it be used? What external data requirements do you have? What changes to the sector will impact on your data processes? What data policies do you already have? Are they up to date and used often? If not, why not? Who is owns the data? Who is responsible for its maintenance What IT/data capability do you have in the organisation? 1/2
- 12. USEFUL TOOLS SWOT analysis PEST/PESTLE analysis The Five Forces Four Corners Critical Success Factors Scenario Planning Value Chain analysis
- 13. QUESTIONS FOR SENIOR MANAGEMENT What questions would you ask if access to data wasn’t an issue? What don’t we know about our students that would be useful? What don’t we know about our staff what would be useful? How do you want the organisation to change over the next 5 years?
- 14. TYPE OF DATA STRATEGY How are your other strategies and policies produced? What approval process does it need to go through? How often do you want to refresh the strategy? Do you use agile techniques in your institution? Who can help you create it? What other departments need to input into it?
- 15. ACTIVITY What other teams will you need to consult with on your data strategy? How quickly can you get institutional approval for the strategy? Will this impact how often you refresh it? Discuss
- 16. DATA STRATEGY DEVELOPMENT At Futureworks I needed approval from the Management Committee, and then the Board of Governors. Initial approval can be gained within a month, final approval takes longer. Changes can’t be made any more often than twice a year Not an agile approach, but suitable for the size of institution and the resource available for changes to be implemented.
- 17. DATA STRATEGY OBJECTIVE What is the objective of your data strategy? Should be a simple statement that sums up how you want the team to act in the medium to long-term. For us: “To build up the data capability of the organisation so that relevant and timely use can be made of internal and external data sources, to support decision making, and ensure the sustainability of the organisation.”
- 18. DATA STRATEGY DEVELOPMENT Link with the organisational mission How does your data capture and use support this? Are you doing anything that doesn’t support the mission? Pick out sections from the mission that you believe can be supported by the data strategy What benefits does your data bring to the organisation? What benefits could it bring?
- 19. LINKING TO INSTITUTIONAL MISSION Two examples: “Academic staff will be given the skills, tools, and capacity to deliver a personalised learning experience for all our students This will be supported by the use of dashboards and analysis to allow tutors to see how their programmes and students are performing. ” “We will regularly review administration processes, business systems and our technological infrastructure – new systems and infrastructure will be required to meet our data needs This strategy will guide how this should be developed and how data will be stored to ensure it meets our future needs. ”
- 20. DATA STRATEGY PRINCIPLES Rules governing behaviour We chose: Data as an asset Data management Data quality Standardisation and linking Accessibility
- 21. DATA STRATEGY PRINCIPLES Data quality “Data should adhere to the principles of accuracy, validity, reliability, timeliness, relevance and completeness. Data will never be perfect but should meet the quality requirements of its intended use, and the quality needs of the external bodies we are required to report to.”
- 22. DATA STRATEGY DELIVERY Our data strategy lists the key areas of development that we need in order to be able to deliver what we want. Data architecture Systems development Business intelligence Organisation & culture
- 23. DATA STRATEGY DELIVERY Organisation and culture “Data will be used to improve operational performance, evaluate options and make better, more sustainable decisions. It is essential that everyone who uses this data understands their responsibilities for maintaining the integrity and quality of our data assets, complying with data legislation and regulations and keeping the data assets safe and secure.”
- 24. DATA STRATEGY LENGTH How long does your data strategy need to be? - Who is going to read it? - How do you want it to be used? - How well are data concepts known in the organisation? - If it’s too long, will people disengage from what you are saying? Long enough to understand, short enough to remember
- 25. ACTIVITY How long do policies in your organisation tend to be? How many of them do people remember? How much training is required to get people doing things as you want them to? How long would you make your data strategy?
- 26. DATA STRATEGY LENGTH Our data strategy is currently: 3 pages + 1 header page
- 27. LESSONS LEARNT Not everyone will think it is a worthwhile task – you may need to spend time explaining the benefits You will never ‘finish’ your data strategy, it should be an ever evolving document, not something you write and then archive Make it accessible to everyone, and reference it any way you can Keep it short enough that people will remember it