Data in business is a commodity and it can be used to highlight areas for optimisation, while also allowing for defining a feasible solution. Over years of engagement with customers from various industries, Opti-Num Solutions has been able to develop a framework for advanced and effective data-driven decision making. Our tried-and-tested approach with the data is broken down into three parts:
The first stage of data driven decision making is to define the problem to be solved. While this may seem trivial, a well-defined problem statement ensures that the investigation for a solution does not venture too far from answering the question. The primary idea is to make the process concise. It also ensures that a good starting point is established for the investigation whereby some initial analytics can be applied to serve as a guideline.
Opti-Num was approached to find a way to increase customer retention through a rewards program for one of our Property Technology (PropTech) clients. With this, Opti-Num was able to define the problem along two components: customer tenure to analyse retention and customer spend to analyse the scale of the rewards.
The second step is to find the cause of the problem, analysing the data and observing the problem to find the underlying cause of problem to be resolved. Looking at the problem’s sources, breaking them down, and examining the individual sources to find the key data sources in a way that then allows the solution to be developed. The process involves delving into the data to find the part of the trend or data group that is causing the problem.
Using the results of the performed analyses, we were able to breakdown customers into distinct categories based on groupings within tenure, spend, and the products they subscribed to. While the data was too distinct, by observing average tenure and spend within each product segment, we could observe the data in more manageable and related groups.
The final stage is to find the solution. After looking through the rest of the data, if there is data that does not suffer from the problem being resolved, we can use it as a base case to mimic. We can look into seeing how we can get the problematic subset to tend towards mimicking the base case. Alternatively, if this is not possible, as the problem permeates all the data, we can use behaviour enforcement to reward the desired behaviour for the user causing a general trend towards resolving the problem.
With this we could identify the average behaviour within each category to define how the rewards should scale with increases in tenure and spend within each product group. The rewards began small and, after a certain amount of time as a customer, began increasing incrementally over time until tenure is just beyond the average for the segment where it then becomes consistent. Changes in their spend would affect the rate at which their reward increases as well.
Our Advanced Busines Analytics team at Opti-Num has successfully been able to define problems using data within various business’, find the root cause of the problems and assist in the development of an innovative solution that can be applied to drive improve behaviour in business. Contact us today to see how we can help take your data and design a solution that is tailored to your business.
What Can I Do Next?
- Request a trial.
- Visit our Advanced Business Analytics Competency Page.
- Find out more from the team.