In our other article on understanding your target customers, we discussed the concept of customer profiling and finding ways to capture more information to build up a clearer picture of your audience. As well a customer profiling, the other exercise an organisation can do to create more marketing intelligence is called customer modelling.
In this case, instead of defining facts and statistics around the type of person/ organisation they are, with customer modelling you instead begin to look at behaviours and try and find a commonality between them. For example, you may decide to try and understand the behaviours that link together some of your highest spending or most loyal customers and understand what differentiates them with those customers who’ve left you. Is there a way of working out a trigger or signal that you can see that highlights which direction someone will take?
The essence of customer modelling is similar in purpose to customer profilings in that you are ultimately trying to build a better understanding and level of intelligence on your market. The core difference in the main approach is that customer modelling is typically more time consuming, particularly depending on how much data you have and how many points it contains.
Typically, customer modelling falls into one of the following brackets:
- Modelling Customer Responsiveness. Defining the characteristics and situational context that makes up the likeliness to respond to a certain situation. Typically this is done by reviewing past data on similar activities and the levels of engagement that made up the performance of that.
- Forecasting Return on Investment (ROI). The more you can understand the expect results of an activity the more you can identity whether it’s going to be both relevant for your audience and if it’s worth spending your marketing budget on. For example, you could look into the historic elasticity of your pricing and promotions and analyse the impacts of adjustments to customer engagement.
- Predicting Behaviour. The more you can understand the aspects that make up a customer’s behaviours and what determines the differences between a loyal and a lost one, the more you can model their future. Customer modelling techniques can significantly aid in calculating the long-term value of a customer to your organisation.
The principles of building a customer model can be as complicated or as concise as you want and are determined by your objectives and the data you hold within your organisation. Keeping a track of things such as pricing calendars, promotional dates, campaign deployment and even governmental initiatives can all prove invaluable.
Experiment with overlaying one on top of the other to try and assess if there are any patterns of trends that determine whether it was a more or less successful period. Use these discoveries to help you determine how to deploy your next campaign, when to run your next promotion or estimate the return from a single customer.