Models
Modeling for Profit
Our modeling premise is straightforward. We use statistical techniques to understand what worked in the past and then project those learnings onto future campaigns.
The classic example is the traditional direct mail response model. Through modeling, we develop an understanding of who is responsive to your offers and who isn't. We then build a model that ranks prospects by their likelihood to respond. Your prospect universe is divided into 10 deciles and a projected response rate is assigned to each decile. The projection will not be a 'guess', as it is based on actual response rates to prior mailings.
But we don't stop at response models. Check out some of the other types of models we build.
Our modeling premise is straightforward. We use statistical techniques to understand what worked in the past and then project those learnings onto future campaigns.
The classic example is the traditional direct mail response model. Through modeling, we develop an understanding of who is responsive to your offers and who isn't. We then build a model that ranks prospects by their likelihood to respond. Your prospect universe is divided into 10 deciles and a projected response rate is assigned to each decile. The projection will not be a 'guess', as it is based on actual response rates to prior mailings.
But we don't stop at response models. Check out some of the other types of models we build.
Sample Modeling Projects
| Profile Models |
Clones your good customer so that you can identify similar prospects. |
| Cross/Up-Sell | Identifies which customers you should offer additional products. |
| Retention | Identifies which customers are most likely to defect to a competitor. |
| Revenue/Value | Predicts a life-time value or projected revenue to each new customer or prospect. |
| Conversion | Predicts which responders are likely to convert, (qualify for your loan, for example, or buy your product). |