Project summary - for my client base pricing mode
Business Problem:
Currently Company X (a real company but name is concealed per company request) charges its client a percentage-of-portfolio size ($) measured in net asset value (NAV) to service and administer client accounts. This has been the pricing model for over 20 years. Recently, Finance ops managers and Investment Managers started hypothesizing that the size of the NAV has little to do with the operational effort (cost to company X) required to service client portfolio. This project seeks to evaluate the hypothesis and either validate the current pricing model or recommend an alternative.
Results:
Decision recommendation - Operating cost per client is driven by other factors other than the NAV. While the NAV amount or portfolio value has some impact on servicing cost, it has the least impact compared to
- Number of cashless securities in the portfolio
- Proportion (%) of non-listed securities. These two factors above have an outsized impact on the daily operational effort needed to service a portfolio, complete pricing, reconciliation, back and middle office activities to validate the end of day/week/month NAV. Regression equation generated to show factors relationship to response (Y) = effort and serve as foundation for new pricing model.
Tech Stack and other tools
- Minitab statistical software
- Manual Data collection – across all operational locations
- Excel
Approach:
Problem definition facilitated workshop to gather all possible cost factors without judgement across 3 continents, data collection across these locations and client type (data collection template attached). The response is cycle time (CT) require completing pricing by staff assessed against the factors from the facilitated session. Finally, regression analysis informed us of 3 factors that impacted operational effort. Regression equation generated to inform new model to price clients.
NOTE: This project was done without code. This was a data project done mostly by hand, data was collected manually (very tedious) guided by a well crafted data collection plan, translated to an easy to understand template for operators in different locations to understand and interprete consistently. Data collected over weeks was then analyzed using Minitab to inform results.