3 Prerequisites for Maximising Value from Data Projects

With recent buzz around AI, many organisations are working to maximise the value from investment into Data & Analytics projects — Some foundational prerequisites should be considered for best outcomes

1. Well-defined Data Strategy

Clear Objectives: Clear definitions of the business problems or objectives you want Data & Analytics use cases to address. The data strategy should include budget and ROI expectations for data initiatives, as well as KPIs and OKRs for individual use cases as well as at an aggregate level.

Alignment to Business Strategy: Data & Analytics strategy with clear links to overall business strategy. Understanding how each use case can enhance your business, whether it’s improving efficiency, customer experience, or decision-making is necessary to apply appropriate focus to the right projects.

2. Data Governance & Management

Data Quality: With inadequate data quality, the output from data projects is at worst case unobtainable and at best case inaccurate. To maximize the value of data projects, ensure that your data is accurate, complete, and relevant. This might involve data cleansing, normalization, and validation processes. To keep data of high quality, a mature data governance setup is required, with Data Owners and Data Stewards taking responsibility for data collected and kept.

• Challenges: Business accountability and prioritization to perform process changes to enhance data quality
• Relevant tools: Soda, Ataccama, Erwing, Atlan

Data Availability: Providing access to relevant data to the business stakeholders drawing value from this data is paramount. Some of the challenges of data democratization can be related to privacy, data security and data proliferation. However, having enough data, especially for projects including machine learning applications, is crucial. If your dataset is too small or unrepresentative, the performance may be limited. Australian company

• Challenges: Privacy, data security & data proliferation
• Relevant tools: Alation, Collibra, Appen (curated datasets for training of algorithms)

3. Data Literate Workforce & Culture

Business Domain Knowledge: Business-specific and functional domain knowledge (incl. data professionals) are essential for identifying relevant use cases, defining problems, and understanding how these can be addressed by data projects.

Standardised Operating Model: Data Projects should be carried out in a consistent manner by teams to allow for cross-leveraging learnings between teams and projects.