Published

How One Hedge Fund Cut Pre-Call Prep From 60 to 5 Min

A hedge fund's analysts packed filings, transcripts, and models into a three-module AI — collapsing pre-call prep from 60 to 5 minutes and reclaiming thousands of hours a year.

80%

Reduced analyst prep, 60 min to 5

4–8 weeks

Implementation Time

Not disclosed

Project Cost

the challenge

A $1 billion AUM hedge fund managing approximately 200 public equities ran an analyst workflow that was bleeding time. Before each corporate access call, analysts spent 30 to 60 minutes per equity manually pulling earnings transcripts, 10-Qs, internal financial models, and prior meeting notes — then synthesizing it all to generate relevant questions. With 15 analysts covering hundreds of equities, the cumulative cost was enormous. The fund had no systematic way to structure pre-call research, post-call transcript analysis, or modeling implications — leaving investment decisions slower, less consistent, and limited by analyst bandwidth.

what they built

Casper Studios mapped the fund’s workflows, data sources, and vendor landscape, then designed a three-module system built around the client’s exact processes. The first module automated Q&A generation, pulling from public filings, earnings transcripts, internal financial models in Excel, SharePoint, and OneNote to produce ready-to-use questions before each call. The second module ingested meeting transcripts and applied sentiment and signal detection to executive responses. The third assessed whether any statement should change buy, sell, or hold assumptions in the fund’s financial models. Each module was shipped independently and iterated on with direct analyst feedback. The unexpected outcome: the fund’s own philosophy — 'singles and doubles over moonshots' — became a guiding framework Casper now applies across all its client engagements.

Casper Studios began by mapping the hedge fund's workflows, data sources, and vendor landscape before designing the system architecture. Rather than building a monolithic tool, they designed three independent modules shipped and iterated on separately, with analyst feedback incorporated after each release. The first module — Q&A generation — pulled from public filings, earnings transcripts, and internal financial models stored in Excel, SharePoint, and OneNote. Before each corporate access call, it produced tailored, ready-to-use questions per equity, cutting prep from 30–60 minutes down to approximately 5 minutes. The second module ingested post-call transcripts and applied sentiment and signal detection to executive responses, reducing time analysts spent processing call output by approximately 50%. The third module assessed whether any statement from the call should change buy, sell, or hold assumptions in the fund's financial models — connecting qualitative conversation to quantitative decision-making. Each module was deployed independently and refined with direct analyst feedback, following the fund's philosophy of 'singles and doubles over moonshots' that Casper has since adopted as a guiding framework.

best fit for

Hedge funds, asset managers, or PE firms where analysts spend disproportionate time on manual pre-meeting research and post-meeting synthesis that can’t scale with headcount — particularly organizations that already use AI individually but haven’t yet connected their tools into workflow-embedded systems.

Ai ROLE

Not shared

impact

80% Reduction in Pre-Call Research Time

The Q&A generation module cut analyst prep time per equity from 30–60 minutes down to approximately 5 minutes, allowing analysts to run multiple preparation queries in parallel and show up to calls better prepared.

50% Reduction in Post-Call Analysis Time

Transcript analysis and financial modeling color modules reduced time spent reviewing call output from roughly one hour to 20–30 minutes per equity, with continued improvement expected as the system matured beyond its first six months.

Thousands of Analyst Hours Reclaimed Annually

Across 15 analysts covering hundreds of equities, the combined time savings across all three modules translated into thousands of analyst hours reclaimed per year — freeing the team to engage more executives and improve investment decision support.

implementation complexity

Not shared

Jay Singh

CEO & Founder

Casper Studios

CEO and co-founder of Casper Studios, a product studio helping companies design, build, and integrate AI-powered products.

Get an intro

Talk to this team

industry

Financial Services

business organization

Finance & Accounting

Operations

AI TYpe

Knowledge Management & Search (RAG)

Data Synthesis & Reporting

Natural Language Processing

value type

Time Savings

Risk & Compliance

other(firm) Case studies

frequently asked questions

How did a mid-sized hedge fund cut pre-call prep from 60 to 5 minutes with RAG and data synthesis?

The mid-sized hedge fund replaced manual pre-call research with a three-module system mapped to its exact workflow. The first module generates ready-to-use questions by pulling from public filings, earnings transcripts, and internal models in Excel, SharePoint, and OneNote; the second applies sentiment and signal detection to call transcripts; the third assesses whether statements should change buy/sell/hold assumptions. Pre-call prep dropped from 30–60 minutes to about 5 minutes per equity.

What AI tools and approach did the hedge fund use?

The system combined knowledge management and search (RAG), data synthesis and reporting, and natural language processing across three independently shipped modules. Tools and platforms included SharePoint, OneNote, Excel, Langfuse, and Wayfound.

What results did the hedge fund achieve?

Three outcomes: an 80% reduction in pre-call research time, from 30–60 minutes to about 5 minutes per equity; a 50% reduction in post-call analysis time, from roughly an hour to 20–30 minutes per equity; and thousands of analyst hours reclaimed annually across 15 analysts covering hundreds of equities.

How long did the engagement take?

Time to results was in the 2–4 month range, with modules shipped and iterated independently and continued improvement expected as the system matured past its first six months.

Who is this AI workflow approach best for?

Hedge funds, asset managers, or PE firms where analysts spend disproportionate time on manual pre- and post-meeting research that can't scale with headcount — especially organizations already using AI individually but not yet connected into workflow-embedded systems.

Have a similar challenge?

Ask whether this would work for you, or describe what you're trying to solve.

TELL US WHAT YOU'RE EXPLORING