Executing a phased data strategy to enable predictive analytics

In mid-2023, a leading authority on risk, undertaking research and sharing practical tools and guidance with its members, turned to Amplifi for help. They are driven by their relevance to their members and their sustainability as a business, with desired outcomes of: 

• Increase memberships & minimise non-renewals 
• Increase revenue from Professional Services
• Ensure relevance of Research, Tools & Events 

The key to success for this organisation, was understanding of how their members, and prospective members, interact with their resources and services and to predict how they will in the future.

The organisation had a vast amount of data related to customers, publications, conferences, and consulting services. However, the data’s reliability and comprehensibility were major issues, hindering their ability to leverage advanced technologies like Generative AI, machine learning, and predictive analytics. 

Amplifi’s primary goal was to develop and execute a transformative data strategy, which would not only enhance their capabilities in analysing member interactions but also enable them to effectively utilise predictive analytics.

Our engagement began with a comprehensive evaluation of the organisation’s needs. They were a small organisation with a small and straightforward systems landscape. As above, they came to us not having much insight into their business but knowing their core desired outcomes they wanted to achieve as an organisation. We first assessed their capabilities to set a baseline from where we could begin. 

At the beginning of this engagement, the organisation’s Data Capability Assessment score was 1.2, signifying a stage ripe for development in the organisation’s data management journey. 

The primary focus at this stage was to analyse member interactions thoroughly to lay the groundwork for our tailored methodology. We soon identified the need for predictive analytics, a key component for forecasting future trends from existing data. Another important discovery was recognising the opportunity to enhance the organisation’s data handling capabilities. We secured the necessary funding for a dedicated data project, aligning the resources with the project’s ambitious scope. 

Designing the strategy 

At this stage of the engagement, we were able to lay out a comprehensive 3-year plan to build the predictive analytics capabilities they desired. This roadmap is structured from the ground up, laying the essential foundations for analytics through improvements in overall data management and the technology landscape. 

The next phase involved crafting the initial version of the data strategy, comprising several key elements: 

• Vision and strategic objectives: We established a clear vision and set strategic objectives to guide the journey of data enhancement. This vision served as a north star for all subsequent initiatives. 
• Capability assessment and gap analysis: We identified the essential capabilities needed to achieve the strategic objectives. A thorough gap analysis was conducted to measure disparities between current and required capabilities. 
• Data management programme: A detailed data management programme was developed, featuring a resourced roadmap with clearly defined short-, medium-, and long-term objectives. This structured and progressive approach ensured that each stage of the journey was well-planned and achievable. 
• Technology platform selection: The selection of technology platforms was meticulously aligned with both business and strategic data needs. This alignment ensured that the chosen technologies would effectively support the desired analytics capabilities. 

 Following this strategic design phase, we assessed again. Their Data Capability Assessment score rose to 2.3, indicating a tangible improvement in the organisation’s data management capabilities. By the end of this phase, the Total Amplifi Time had reached 3 months, marking the completion of the strategic foundation and setting the stage for the upcoming implementation phases. 

Phase 1 – Piloting the strategy 

This phase involved establishing a data organisation structure, setting up a data management and governance operating model, and implementing modern analytics architecture. Data and metadata processes were defined and piloted. 

After we implemented the strategy pilot, we assessed again. Their Data Capability Assessment score increased to 4, signalling significant data governance and quality improvements. The total time for this phase was 8 months. This is evidence to why it’s essential to assess, assess, assess. You don’t know if you’re heading in the right direction without it. 

During our Pilot, we were able to prove that applying good practices to specific situations can yield successful results. These successes allowed us to then roll out these practices across various business domains, continuing along the path of the developed strategy roadmap. However, we’re mindful that the data landscape is constantly evolving. The growing ease of using Large Language Models and GPT has introduced new possibilities, especially in enterprise search. Our strategy’s inherent flexibility not only allows us to establish a solid foundation but also enables us to adapt and embrace new technologies and opportunities as they arise. 

Phase 2 – Rollout and continuous innovation 

Following the successful pilot, we embarked on a broader rollout, implementing our refined operating models and governance across various business departments. This step was essential for integrating the pilot’s successes and learnings into the wider organisational framework. Throughout this phase, both the strategy and analytics operating models were continuously refined, including significant advancements in cloud operations and AI exploitation. 

An essential element of our approach was the recognition of the potential limitations of a rigid, top-down strategy. If the focus is solely on delivering what senior leadership wants to achieve, there’s a risk of missing out on unforeseen opportunities along the way. A successful strategy is about focussing yourself in the right direction; however, it’s also about allowing room for innovation and the ability to seize golden opportunities as they arise. This approach ensures the process isn’t merely mechanical but dynamic and adaptable. 

To counteract this, we placed a strong emphasis on innovation as a core component of our strategy, ensuring flexibility. This allowed the organisation to adeptly adapt to emerging technologies, particularly in the use of Large Language Models and GPT. These have been applied to enterprise search and to predicting how proposed research publications will perform based upon the historical performance of semantically similar pieces. 

Implementation of this strategy will ultimately result in improved data reliability, actionable insights through predictive analytics, and enhanced user interactions via a private LLM application.

These improvements have already directly contributed to the strategic objectives of minimising non-renewals, increasing membership, and ensuring the relevance of the organisation’s research and services. The methodology’s tangible and practical approach has seen the significant improvement in Data Capability Assessment scores, which highlights the value of continued assessment and understanding your starting point as an organisation when approach data strategy.