Earnix Blog > Customer Centricity
Dynamische Preisgestaltung in Echtzeit im Bank- und Versicherungswesen
Liat Mendelson
20. November 2018
- Customer Centricity
Real-Time Price Tuning to Fit Dynamic Markets
Modern financial services markets are highly dynamic. Prices change often – due to seasonality, changes in competitor pricing and other marketplace variables. However, developing the best pricing strategy for insurance and banking products can take months.
When you consider the time needed to wade through local regulatory approvals, IT development of the pricing models, and several other organizational processes, prices can often become stale. As a result, when finally reaching the market, they are no longer competitive, and as a result, KPIs are not able to be reached.
What if your pricing algorithm could automatically fine tune the prices to reach organizational target KPIs?
With Earnix Price-it, sensitive price changes can be made to reflect new market conditions. Minor adjustments to prices, in defined thresholds, allow insurers or banks to ensure that their market prices truly reflect current market conditions. Differences between the actual and predicted KPI (e.g. market conversion, profit etc.) can be detected and the Earnix pricing software can automatically change the price – in order to allow a better fit to target KPIs.
How will insurers and banks benefit from this algorithm?
• Faster market response. Prices can be fine-tuned at any point, particularly when a difference between predicted and actual KPIs are observed. There is no need to recreate models repeatedly.
• Stay aligned with KPIs. The Earnix algorithm can constantly monitor KPI levels and make required changes to keep prices on target.
• Efficient pricing saves time, money, and resources. With automatic changes made to pricing models and versions, no time is spent building, testing, approving, filing and deploying net new models to online environments. Pricing changes are completed in an automated fashion in several minutes, so that analysts can spend time focusing on strategic activities of the business.
Real Time Price Tuning in Action - How it Works
If we look at the graph below, we have a data sample of 24 periods with target conversion rate of 14% set as our KPI. Starting at week 14, we can see that there is a significant gap between the market behavior and the target conversion. The model prediction doesn’t fit the actual market behavior.
Using the Real-Time Price Tuning algorithm, Earnix can update the model parameters to allow better fit to the market behavior and thus provide a better prediction. In the example below, where the target KPI is the conversion rate, the only data that is needed to employ this methodology is the difference between original predicted demand and actual average demand. This can be gathered based on recent weeks data. Once adjusted, the model will provide a better prediction and thus improved business results. These adjustments can start from the first day the model is deployed, not necessarily from when the drop off occurs in week 14!
Putting Banks and Insurers in Control.
With the ability to set thresholds, or a range by which the system can act, no changes will be made beyond the defined threshold. Organizations can monitor actual results compared to model predictions to ensure the real-time price tuning is reasonable, acceptable and meeting the target KPIs originally set. If an organization does detect significant changes, there will be a need for a new model and pricing strategy to be created. Once this is complete, new prices can be set according to the new model.
What’s next?
With these pricing tuning algorithms currently in development, we are now focusing on automatic model creation and deployment. The Earnix roadmap includes capabilities to automatically update models, thereby enabling banks and insurers to deploy new models automatically, not just tweak the existing model parameters within defined thresholds. This new methodology will detect changes in external variables, such as population mix, as well as significant changes in target KPIs moving them out of a defined threshold, and automatically build a new optimization model to better fit actual market behavior. Stay tuned for more on this exciting new technology coming from Earnix!
Modern financial services markets are highly dynamic. Prices change often – due to seasonality, changes in competitor pricing and other marketplace variables. However, developing the best pricing strategy for insurance and banking products can take months.
When you consider the time needed to wade through local regulatory approvals, IT development of the pricing models, and several other organizational processes, prices can often become stale. As a result, when finally reaching the market, they are no longer competitive, and as a result, KPIs are not able to be reached.
What if your pricing algorithm could automatically fine tune the prices to reach organizational target KPIs?
With Earnix Price-it, sensitive price changes can be made to reflect new market conditions. Minor adjustments to prices, in defined thresholds, allow insurers or banks to ensure that their market prices truly reflect current market conditions. Differences between the actual and predicted KPI (e.g. market conversion, profit etc.) can be detected and the Earnix pricing software can automatically change the price – in order to allow a better fit to target KPIs.
How will insurers and banks benefit from this algorithm?
• Faster market response. Prices can be fine-tuned at any point, particularly when a difference between predicted and actual KPIs are observed. There is no need to recreate models repeatedly.
• Stay aligned with KPIs. The Earnix algorithm can constantly monitor KPI levels and make required changes to keep prices on target.
• Efficient pricing saves time, money, and resources. With automatic changes made to pricing models and versions, no time is spent building, testing, approving, filing and deploying net new models to online environments. Pricing changes are completed in an automated fashion in several minutes, so that analysts can spend time focusing on strategic activities of the business.
Real Time Price Tuning in Action - How it Works
If we look at the graph below, we have a data sample of 24 periods with target conversion rate of 14% set as our KPI. Starting at week 14, we can see that there is a significant gap between the market behavior and the target conversion. The model prediction doesn’t fit the actual market behavior.
Using the Real-Time Price Tuning algorithm, Earnix can update the model parameters to allow better fit to the market behavior and thus provide a better prediction. In the example below, where the target KPI is the conversion rate, the only data that is needed to employ this methodology is the difference between original predicted demand and actual average demand. This can be gathered based on recent weeks data. Once adjusted, the model will provide a better prediction and thus improved business results. These adjustments can start from the first day the model is deployed, not necessarily from when the drop off occurs in week 14!
Putting Banks and Insurers in Control.
With the ability to set thresholds, or a range by which the system can act, no changes will be made beyond the defined threshold. Organizations can monitor actual results compared to model predictions to ensure the real-time price tuning is reasonable, acceptable and meeting the target KPIs originally set. If an organization does detect significant changes, there will be a need for a new model and pricing strategy to be created. Once this is complete, new prices can be set according to the new model.
What’s next?
With these pricing tuning algorithms currently in development, we are now focusing on automatic model creation and deployment. The Earnix roadmap includes capabilities to automatically update models, thereby enabling banks and insurers to deploy new models automatically, not just tweak the existing model parameters within defined thresholds. This new methodology will detect changes in external variables, such as population mix, as well as significant changes in target KPIs moving them out of a defined threshold, and automatically build a new optimization model to better fit actual market behavior. Stay tuned for more on this exciting new technology coming from Earnix!
Teilen article: