e-news

What is the future for optimisation technology?

In the search for cutting edge innovation, optimisation software already has the potential to bring real value to businesses looking to maximise performance.

However, the next generation of the optimisation technology is likely to bring to market a range of new features that will add even more power to the science of decisions.

Optimisation is defined as an act, process, or methodology of making something (as a design, system, or decision) as fully perfect, functional, or effective as possible. There are many areas that businesses can utilise optimisation software. Throughout the lifecycle the technology can be used to improve performance.

Specific areas include:

• Prospecting - Targeting, maximising market share and profitability from new customer acquisition campaigns. i.e. Which customers to target, which offer.
  
• Origination - Registration, improve risk-adjusted return on assets. i.e. Which customers to approve, what initial limits and terms to offer.
  
• Customer Management - Maximise customer value from marketing campaigns. i.e. Which customers, how often to contact, which offer.
  
• Collections Recoveries - Improve collections and recovery values. i.e. which accounts to collect, which treatment to apply.
  

Driven by client requirements for efficiency and delivering tangible results quickly, augmentations such as Impact Analysis and Scenario Analysis will bring robust additional features that will deliver added value.

Scenario Analysis
Scenario Analysis is the ability to derive the predicted results given user-defined constraints, goals and conditions (e.g. how much more or less profit can I generate with a budget of 'X' as opposed to 'Y').

Scenario Analysis will also allow the ability to assess the quality of models by analysing model predictions versus actual performance.

Impact Analysis
Impact Analysis will provide an additional capability to evaluate the potential of overall strategy performance results by considering improvements to the existing models.

Impact Analysis will enable streamlining of analytical efforts to optimise strategies. The key question it addresses is ‘What-IF modelling’. The result of this is the ability to estimate ‘what benefits would I see if my model improved by a factor of x?’  This question can now be answered quickly and with zero development efforts.

Optimisation users with access to Impact Analysis will be able to accurately estimate the impact of improving the predictability of a model, without actually having to build the model.  So, in a decision optimisation scenario we can say for instance: ‘If you rebuild model x we can increase your profitability by $300K’ and this can be demonstrated in the software with no development effort.

For both credit and marketing optimisation problems representing customer features / behaviour accurately is the key to ensuring the correct customer decisions are made. The next generation of optimisation technology will allow an instant assessment of the impact of enhanced models on a portfolio.

Practically optimisation users will be able to follow a basic approach:

• Start with a basic simple model, such as a portfolio with an average response model to offer A.
• Decide if improving this model would generate higher value added by making better optimised decisions.
• Assess models on an individual basis and test the impact on your overall strategy

OR

• Assess the impact of a combined set of models on overall strategies

Impact Analysis will offer the ability to have increased flexibility and decision making, this will include:

1. Insight into which model is critical to your overall business goal
2. A new approach for creating your analytical roadmap
3. The capacity to combine your roadmap with a ROI calculator to enable prioritisation of model development
4.  Impact Analysis will reduce the lead-time to model development and a test and learn approach can be reduced to one step.
5. Reduce the cost of (re-)developing useless models

The conclusion is clear overall, the next generation of optimisation technology, with features such as Impact Analysis, will help add a significant amount to the revenue enhancing potential of optimisation software.

Delphine Luong
Global Consultant
Decision Analytics
Experian

If you would like to know more about this subject or have any questions for the author please Contact us