Having looked at important considerations in Customer Lifetime
Value (CLV) model design, this second blog in our CLV series looks
at the critical factors in founding a stable platform for your
model. To successfully operationalise a CLV model, a comprehensive
operating model is required, considering people and organisation,
data and governance, and technology. These contextual factors are
often overlooked in the build of a model and become the roadblock
to the realisation of the benefit from even a robust model. We will
look at each of these areas in turn.
1. People and organisation
CLV can realise benefits across a business but with each function having unique requirements, the identification of key stakeholders and their requirements can be challenging. Becoming a CLV-driven organisation requires clarity around how each function will use CLV to enhance or develop existing processes. Functions often require guidance on the possibilities CLV provides, from prioritised acquisition, retention and growth strategies to product development and capital investment decision-making. This raises a number of important questions:
- Can one model fit the needs of all our teams?
- Who should own and deliver it to those teams?
- How do we keep models and outputs controlled and aligned?
The biggest challenges we see tend to involve models that are semi-developed and the centre of disagreement between functions due to different and misunderstood requirements. Alternatively, different teams develop their own irreconcilable models which do not achieve similar outcomes. Often this stalls development and implementation of the model, undermining senior support, and resulting in wasted effort and unrealised potential (this will be examined further in the final instalment of this series). Focussing on the following can help avoid these issues:
- Identify and engage core stakeholders
- Agree scope, gather and communicate clear requirements, understanding how outputs differ by stakeholder group
- Define a core owner of the model and governance, with RACI defined for all stakeholders
2. Data and governance
Central to engendering confidence in the model is building a stable base and consistent, reconcilable outputs. Control of data and governance is core to this stability:
- Governance is non-negotiable and must be policed strictly by the model owner – preferably the owner will be independent from the stakeholder groups, such as a dedicated analytics function.
- CLV data input requirements need to be understood by relevant operational areas to provide an understanding of the granularity and accuracy available to the model, and the impact this has on outputs.
- Clear definitions on data period and provision of data to stakeholders across the business to ensure that data is available to the different business users in time and in the right format.
- Although different models might cater to different teams, all models should share common data and definitions. To establish and maintain a single version of the truth helping to reduce the risk and costs of data storage.
3. Technology
Technology is important for the creation, management, maintenance and use of the output from a CLV model. An organisation's analytical and technological maturity, budget and CLV requirements will be defining factors regarding accuracy, management and use of the model. Suitable technology can drive full realisation of the benefits from a CLV model, but the key is matching technology to your needs and means as a business. Complex processing capability and statistical platforms can enhance granularity and accuracy, but are dependent on quality inputs and knowledgeable teams to operate. Often simple Microsoft Excel models based on small amounts of data can provide real progression at minimal cost. That said, always consider scalability of your platform to potential requirements to avoid foreseeable re-platforming due to growth in scope or requirements. Technology is similarly key to effective application of the CLV model outputs, to be discussed in the final instalment of this series.
Conclusions
The implementation of a stable platform for your CLV model is dependent on a comprehensive and controlled operating model encompassing all of the above. However, a consistent theme throughout this blog series is that the eventual guise of this is dependent on the needs and means of your business, and CLV can be implemented successfully at many varying levels of complexity. Central to success is focussing on certain guidelines:
1. Agree a central, independent
owner for the model and provide them with the means to achieve
their goal
2. Define a clear scope
3. Maintain clear alignment between models from the start and
make intelligent decisions on where the benefit / complexity
cut-off lies between and within models
4. Lock down common inputs, definitions and governance of the
models
Our final instalment in this series considers the complexities and decisions involved in the effective application of CLV outputs with a view to fully realising the benefits of the efforts involved so far.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
