Execution at the front line: realising the value of strategy analytics

Set a clear business vision for the use of strategy analytics

Laying down a clear vision for the use of data that is intrinsically tied to the overall company goals is essential to establish business results as the number one priority. Prioritising the business goals also helps ensure that it’s never a case of implementing analytics for analytics’ sake. So, before asking what the data can do for you, identify the business questions where data insights can really add value. Be specific about the key data metrics that will inform strategic choices and, ultimately, facilitate actions to drive business growth. Define and communicate in a tangible way for sponsors and stakeholders what you want to achieve with that data – e.g. reduce cost of customer acquisitions by 20%. Find the leaders and key stakeholders who will act as change champions and communicate this vision for a new data-driven approach. Tactically speaking, starting ‘small’ and scaling with the business area that will likely yield the biggest value from analytics and where data availability and receptivity is highest can be key to gain early momentum in executing the business vision.

Whilst a company-wide transformation or integration of data scientists across the business might be the eventual goal, the immediate focus of execution should be on proving the business value of analytics to the organisation by delivering concrete results. First establish a proof of concept, or use case, to address a key strategic question in a specific business area and construct a leading hypothesis and metrics to test.

Alongside defining, testing, and proving the model ability (predictability or optimisation), involve business counterparts early on in the process to develop the analytics solution. Specifically, business owners should inform and prioritise the key strategic insights required, validate assumptions used in the methodology, set the granularity, contextualise the model with business logic, and highlight challenges in functionality or user experience, before scaling the analytics to a wider audience. Creating a core driving team, constituted of data scientists as well as business owners, will help to align the analytics development with the core business needs.

Executing in sprints will help to prioritise model development needs and business questions to address, as well as enabling a quick processing of learnings and releasing of a minimum viable product to test with the business. With success from a proof of concept (e.g. increased sales revenue or higher customer conversion) realised on a micro scale (e.g. at a region or neighbourhood level), the model can be gradually expanded, to address additional prioritised business questions or more regions or customers.

At our own clients, we approach the journey towards establishing a data-driven strategy by not only defining the analytics strategy but also by quickly proving the value of analytics - prioritising proof of concepts to experience the value of generated analytics solutions.

One example is the work that was done for a global manufacturing company. Next to helping them define the strategy for the next two years to grow their analytics capabilities and ambitions, the client gained an understanding of the business value that analytics could deliver through the insights gained from three proof of concepts:

1. Predicting false alerts for sensors, with a 79% accuracy
2. Segmenting fleet managers into seven distinct segments based on their behaviour
3. Using statistical modelling to understand key failure factors for machinery.

The value that these proof of concepts created helped to generate momentum for further investment in analytics capabilities. We have seen that choosing a narrower scope with respect to regions or functions initially and proving the value of analytics helps to more easily expand these initiatives to other regions and functions at a later stage. Another example is a global retail company, where Deloitte started off with proving the value of analytics with a successful customer segmentation, and then expanded to optimising product assortments based on these customer segments.

Empower employees with the knowledge to make informed strategic decisions

In this blog, we focused on the importance of defining a vision for strategy analytics and executing through establishing proof of concepts to realise the value of analytics before scaling gradually. Empowering employees with analytics also requires engaging the business in shaping the process, to ensure models are designed in such a way that they enhance the decision making of owners. Finally, our closing assertion is that one – Strategy – does not go without the other – Analytics; taking decisions about the future requires defining strategic choices and selecting those which present the strongest argument. In assessing which has the strongest argument, analytics should be consistently used to provide concrete evidence, alongside business judgement, imagination and the experience of employees.