In an outbound sales environment, the agent needs to work through a long list of customers and the more information available to the agent on the customer, the better.
We’ve identified two types of models that can assist the agent and improve outbound campaign results.
Models to use in a sales call centre
Take-up model:
A propensity to take-up model, allows an agent to understand the likelihood of a client being interested in taking up the offer. It is typically used to prioritise leads, but it is also useful information for the agent when phoning the customer. The agent can adjust his conversation accordingly and be more prepared for the call if he knows that the customer will be more or less likely to be interested.
Best Product model:
The second model is the best product model. This model can be used to identify which product the customer is more likely to be interested in. The model will also give you the probability to take-up the offer, similar to the previous model. It can also be used to identify the second or third best product to offer. Therefore, it basically gives you a product ranking and likelihood to take-up.
Methodology and benefits
The models developed will predict how likely a client is to be interested in the specific offer or product. The model uses past behaviour and trends to predict future behaviour. Therefore, it is essential that the historic campaign data is available for the same offer or product. The model will allow you to rank the universe according to the likelihood of taking up the offer.
If you have multiple products and are interested to know which products to offer to which customers, the best product model can be used. This methodology gives you a model per product that predicts the likelihood of a customer taking up the product. These models can be used to identify the best product and also rank the products per customer based on likelihood to take-up.
Data used
To develop a propensity to take-up model, the following historical data sources are required:
- Dialler data indicating who was contacted together with the dialler outcome
- Sales data indicating who took up the offer
- It is essential to be able to match the sales and dialler data on the unique customer ID
- Bureau data to be used as predictive characteristics for SACRRA members
- Other demographic and geospatial data to be used where the client is not a SACRRA member
To develop a best product model, the following historical data sources are required:
- Dialler data indicating who was contacted together with the dialler outcome for each of the products
- Sales data indicating who took up for each of the products
- It is essential to be able to match the sales and dialler data on the unique customer ID
- Bureau data to be used as predictive characteristics for SACRRA members
- Other demographic and geospatial data to be used where the client is not a SACRRA member
Keep your agents informed and empowered
It is vitally important, however, that your call centre agents have easy-access to the outcomes and recommendations made by these models.
With models built by Principa in our proprietary decisioning software, DecisionSmart, Principa’s Agent X Call Centre Virtual Assistant can deliver real-time call and agent specific data, decisions and information to your agents visually, to inform, guide and motivate them.
