Save the Quants

Inspiration

Explaining complex portfolio performance to investors without a financial background can be difficult and time-consuming.

Our research showed that simply giving investors more financial information does not necessarily mean they will understand it correctly. Even when reports are available, complex information can still be difficult to interpret, particularly for people with less financial experience. This can lead to misunderstanding and make informed decision-making harder.

This made us ask a broader question: does the same communication gap exist in quantitative investment firms, where professionals need to explain complex portfolio results to investors?

We found that investors need to understand what happened to their portfolio in order to meaningfully evaluate an investment or manager. But producing clear explanations for different investors can also become difficult to scale.

This inspired us to build Save the Quants: a portfolio translator designed to make quantitative performance easier to understand.

What it does

Save the Quants turns daily portfolio performance data into explanations tailored to different audiences.

Users upload portfolio data, select a benchmark and audience, and receive quantitative insights, charts, and a report written at an appropriate level of technical detail.

The underlying financial facts stay the same. What changes is how those facts are prioritised and explained.

How we built it

We designed the system around three stages:

Facts -> Relevance -> Communication

First, the system calculates portfolio performance and risk metrics using deterministic code.

Next, it identifies and prioritises relevant insights for the selected audience.

Finally, AI turns those findings into an explanation suited to the reader's level of financial knowledge.

This allows us to use AI for communication without relying on it to calculate the underlying financial results.

Challenges we ran into

One challenge was translating our research into actual product decisions.

It was not enough to simply make financial language "simpler." Different audiences may need different information, different levels of detail, and different explanations.

We also had to decide what the system could reasonably conclude from daily portfolio values. The data can tell us what happened — such as return, volatility or drawdown — but it cannot always explain why it happened.

This led to one of our key principles:

Don't infer what the data doesn't tell you.

Accomplishments that we're proud of

We are proud that we turned a research-backed communication problem into a working prototype.

The same portfolio can be analysed once and then communicated differently depending on who is reading it, while keeping the underlying financial facts consistent.

We also created a complete workflow from raw portfolio data to quantitative analysis, audience-aware insights, explanations, charts and a final report.

Our goal is not to replace quantitative professionals.

We don't replace the Quant. We scale the Quant.

What we learned

The biggest thing we learned is that access to financial information is not the same as understanding it.

Effective communication requires thinking about who is receiving the information, what is relevant to them, and how much technical detail they need.

We also learned that AI can be most useful when it complements structured analysis rather than replacing it. In our system, quantitative code establishes the facts, while AI helps communicate those facts more effectively.

What's next for Save the Quants

The next step is to validate the system with more real users and finance professionals to better understand what different audiences actually need from portfolio reports.

We would also like to support more audience types and richer portfolio information, such as holdings, transactions and factor exposures.

With richer data, the system could eventually move from explaining what happened to providing stronger evidence about why it happened.