In the second article of our three-part series, we explore the pivotal role that models play in answering specific types of questions and the critical importance of integrating business insight to make these models strategically valuable.
THE STRENGTH OF MODELS IN PREDICTIVE DECISION-MAKING
Models are incredibly effective tools for answering certain types of questions, especially when it comes to predictive decision-making. Unlike historical reporting, which tells us what has happened, models help us forecast how today’s tactical decisions will influence tomorrow’s strategic objectives.
At their core, models are mathematical constructs that quantify relationships between a set of inputs (independent variables) and an output (dependent variable). For example, a model might address the question, “How will our conversion rate change as our price increases, and is the response different for single- versus multi-product customers?” This type of question has clear drivers (price and number of products) that impact an outcome (conversion rate).
MODELS AS DECISION SUPPORT TOOLS
It is essential to remember that models assist with – but do not replace – decision-making. They are constrained by the input data and the questions they are designed to answer. For instance, a model built for price optimization of cash acquisition offers is not likely to assess the impact of whether the optimal price may alienate certain customer segments. Therefore, the strategic value of a model is significantly enhanced when it is integrated with business insights.
THE ROLE OF BUSINESS INSIGHTS
Each of us uses heuristics to understand how specific events impact outcomes: deposit rates rise when the Fed Funds Rate increases, mortgage applications decrease when mortgage rates go up, customers attracted by a teaser rate leave after the teaser expires, and higher-risk customers are more likely to default.
As we gain experience, these heuristics become richer and more nuanced. For instance, deposit rates might change at different speeds depending on whether the Fed hikes or cuts rates, and mortgage applications might drop when a bank’s mortgage rate becomes less competitive.
The best models quantify the strengths of these relationships. Just like our mental heuristics, models become more accurate with deeper experience and more data. Sometimes, this process even prompts us to re-evaluate our beliefs.
DIFFERENT MODELS FOR DIFFERENT QUESTIONS
Various modeling forms are designed to address different types of questions. OLS (Ordinary Least Squares) models predict amounts, ARIMA (AutoRegressive Integrated Moving Average) models forecast time series, and logistic models estimate probabilities. Developing these models is a collaborative process between business specialists and model specialists. The business specialists know the right questions to ask, and the model specialists translate these questions into mathematical frameworks.
NOMIS' APPROACH TO PRICE OPTIMIZATION
At Nomis, we specialize in the science of price optimization. A key aspect of our approach is to continuously interrogate and respond to the price and non-price features that influence price elasticity. This methodology allows us to tailor our insights to the specific level of pricing sophistication of each financial institution we work with, ensuring the most accurate and actionable insights possible.
Models are powerful tools that support predictive decision-making by quantifying complex relationships between variables. However, their true strategic value is realized when they are combined with deep business insights. By understanding the nuances of these relationships and continuously refining our models, we can provide financial institutions with the precise, actionable insights they need to optimize their pricing strategies.
Don’t miss the third article of our series, where we will discuss the importance of not letting failed statistical tests derail useful models and the necessity of planning for model failures.
For more information and to access the full series, visit Nomis Solutions.
About the Author:
Wes West is the Chief Analytics Officer at Nomis Solutions, the foremost provider of end-to-end pricing analytics and execution technology. With a distinguished career at leading financial institutions, West brings unparalleled expertise and a track record of innovation to his role at Nomis. His extensive experience spans Retail, Strategy, Finance, and Treasury, offering unique insights into the world of financial analytics and strategic pricing. Connect with him on LinkedIn or reach out via email at wes.west@nomissolutions.com.
