Industry
Financial Investments
Use Case
Pricing Optimization
The Challenge
Mutual fund managers face a surprisingly complex pricing problem. Unlike most retail products, a mutual fund doesn't have a single price — it has a layered set of fees, including front-end loads, expense ratios, 12b-1 fees, and more. Each of those levers affect both revenue and how attractive the fund looks to potential investors. Getting that balance right is hard, and the stakes are high.
Traditional approaches to fund pricing leaned heavily on financial modeling or competitive benchmarking, but they didn't account for how investors actually makechoices. Fund managers knew that fees mattered, but they didn't have a clear way to quantify the trade-offs consumers were making or identify which fee structures would hold up best across different investor segments.
The Solution
Wilcox built a management decision model that pairs revenue projections with consumer preference data collected through conjoint analysis. The approach works by mapping out "iso-revenue curves", the many different fee combinations that generate the same expected revenue from a given investor, and then overlaying those with utility scores from the conjoint. The fee structures sitting at the intersection of high revenue potential and strong consumer preference form what the model calls an "efficient frontier."
To go deeper, Wilcox used latent-class conjoint analysis to segment investors into groups with meaningfully differentiated decision-making patterns. One group cared primarily about fees; the other focused almost entirely on past performance. Each group pointed to a different set of efficient fee structures, opening the door for mutual fund companies to tailor their offerings to different investor profiles.
Sawtooth tools and methods used:
- Latent-Class Choice-Based Conjoint (LC CBC) — captured how different investor segments weigh fee structures when choosing between funds, enabling segment-specific optimization
- Part-worth estimation and linear interpolation — translated conjoint outputs into utility scores for any fee combination within the tested range, feeding directly into the optimization model
The Outcome
- A practical tool for pricing decisions. The model rules out a large portion of possible fee structures as inefficient, narrowing the decision space to a much smaller set of strong candidates. Fund managers can then apply their own competitive and strategic knowledge to choose from that shortlist.
- Segment-specific fee guidance. Latent-class analysis revealed two meaningfully different investor groups. Fee-sensitive investors favored higher loads paired with lower expense ratios. Performance-focused investors were largely indifferent to fee structure details, suggesting expense ratios could rise without significantly affecting their fund preference.
- A foundation for smarter price discrimination. By identifying which investors belong to which segment, fund companies can design and target different fee structures towards different audiences, capturing more value across the market rather than settling for a single one-size-fits-all approach.
- Validated by real investor behavior. The model was demonstrated using choice data from 50 current mutual fund investors across 20 choice tasks each, confirming that investor preferences for fee structures don't always align with pure cost minimization — a critical finding for managers who assumed otherwise.