How can we get more $$$ out of our existing customer?
Across different industries pricing increases for existing customers happen in a variety of different ways. The three most common ways I have come across are:
- Blanket % increase based on ‘Inflation’ or ‘Increase in fuel costs’.
- Customer specific increases based on a gut feeling that they would be happy to pay a little bit more.
- Roll out of pre-determined yearly contractual price increases.
This currently works but could we be a bit smarter?
Deep within a Companies Terms & Conditions of many if not all of their customer product and service offerings is a single line:
“We reserve the right to increase the price at any time.”
So why don’t we execute this right? – If we don’t are we not we leaving $$ on the table?
The biggest fear is usually around customer retention but what if there was a way to monitor each and ever customer and based on our history with them predict how much more we could potentially squeeze out of them?
Well that is where Machine Learning could assist
Azure Machine Learning Module: Survival Analysis API
Brief description: Survival Analysis API provided by Microsoft allows us to provide the machine learning tool with a series of inputs and for the tool provides us with a date when a certain event is likely to occur – For us in this scenario would be Customer Termination
Knowing this this date we are able dynamically manipulate a series price changes and other retention methods to get as much out of the pockets of customers while also retaining them as a customer.
Possible data inputs: Depending on industry there will be different ways of measuring if a customer is happy to pay more below are a few generic inputs which could be inputted into the tool.
- Historical customer pricing increases including pricing and retentions.
- Number of customer support tickets and resolution times (Assess customer satisfaction).
- Customer feedback (Assess customer satisfaction).
- Any other consistent business metric…
Many companies are likely to be leaving thousands if not millions of dollars within existing customer pockets. With the help of machine learning we could measure specific KPI’s and historical trends to individually assess each of our existing customer and provide a customer specific recommended price increase while also retaining them as a customer.