Hi everyone,
This edition introduces two of the strongest global workflows we have built so far using Grok 4.1.
Both are universal.
Both work for India, US, Europe, APAC or any listed company.
Both are designed for analysts, PMs, advisors, students and anyone who relies on deep fundamental analysis.
What is inside this newsletter
You will receive:
Global Equity Research Coverage Prompt
Global Earnings Call Tone Shift Dashboard Prompt
Reference post and output from our Krsnaa Diagnostics test
Interpretation guidelines for both workflows
Market use cases
Access instructions
This edition is focused on one idea:
AI can now support end to end research workflows at a level that analysts can actually use.
1. Global Equity Research Coverage Prompt
A full scale research prompt that produces a complete, structured coverage style report for any global company.
Built for both emerging markets and developed markets.
The prompt covers:
Key data snapshot
Thesis
Drivers
Operating assumptions
Peer comparison
Valuation
Risks
Clean appendix structure
During our test with Krsnaa Diagnostics, Grok 4.1 produced a report with reasoning, sector alignment and clarity that felt far beyond typical LLM outputs.
This prompt works for any country and any market. You can specify the research house style you want, for example Goldman Sachs or Morgan Stanley or Citi. The structure adjusts automatically.
It works the same for Nvidia, Tesla, Adani Ports, Toyota, ASML, Reliance, Samsung, or any global listed company.
Required Inputs
To run this prompt, users must provide four inputs:
Company name
Ticker
Exchange
Country
Preferred research house style
(For example: Goldman Sachs or Morgan Stanley or Citi or UBS)
The prompt will reconstruct the entire report around these inputs.
This has become our base workflow for analysts who want to test coverage frameworks across markets.
2. Global Earnings Call Tone Shift Dashboard Prompt
This is the second workflow.
In many ways, it is even more powerful.
The prompt reads two earnings call transcripts and builds a fully responsive HTML dashboard that highlights:
Tone shifts
Guidance changes
Confidence vs caution
Operational commentary
Financial signals
Additions and deletions across quarters
Sentiment heatmaps
Q and A surprises
Capital allocation language
Positives and red flags that could move the stock
The dashboard is structured for institutional use:
Key components
• Message Shift Map
Interactive diff showing what management added, removed or softened between quarters.
• Six Panel Executive Summary
Macro Pulse
Demand Outlook
Margins
Guidance Tone
Q and A Surprises
Capital Allocation
• Tone and Intent Radar
Plotly based sentiment heatmap for each speaker block.
• Metric Tracker
Extracts all financial and operational metrics
Revenue, margins, volumes, AUM, NIM, PLF, backlog, capex, cash, debt
Flags metrics with five percent or more movement
Marks missing numbers as N A automatically
• Investor Playbook
A short note on what matters next for the stock.
• Full HTML export
Responsive, clean blue toned layout
Sticky navigation
Cards, highlights, and section anchors
Footer tagging as “Generated by Grok Studio”
Why this matters
Earnings calls are where management reveals intent.
But the signals are subtle.
Language, tone, omissions, guidance shifts.
This prompt gives you:
Real quarter vs quarter comparison
The exact sections where tone changed
Operational signals for near term stock impact
A dashboard you can store, compare or share
This is one of the strongest management insight tools we have built.
Reference Demo
We tested both workflows this week to understand how far Grok 4.1 can be pushed in real research scenarios.
Equity Research Demo: Krsnaa Diagnostics
Output Results: View
Earnings Call Tone Shift Demo: Reliance Industries
Output Results: View
How to Use These Workflows
Both prompts are global.
They work for:
NASDAQ
NYSE
LSE
HKEX
SGX
NSE or BSE
Tokyo
Frankfurt
Any earnings call transcript in PDF, DOC or TXT
For the equity research prompt, enter:
<Company_Name> (example: Nvidia or Reliance or Toyota)
<Exchange> (example: NASDAQ or NSE or HKEX or LSE)
Add Country
Preferred bank style
This ensures the model produces a complete, globally structured coverage report in the chosen style.
For the Global Earnings Call Tone Shift Dashboard prompt, enter:
the prompt.
Upload the two transcripts.
Add the company name inside Prompt.
Grok handles the rest.
Execution time: 2 to 5 minutes depending on transcript length.
Our 3 Tool Data Integrity Workflow for Institutional Accuracy
One reason our outputs were so strong this week is that we did not rely on one tool to do everything.
Most people use AI incorrectly. They ask a single model to read filings, fetch data, interpret numbers and build analysis. That is how inconsistencies appear.
Our workflow follows one core rule.
Data integrity first. Synthesis second.
Here is the system we used:
Step 1: NotebookLM
We uploaded the full annual report into NotebookLM to extract clean, grounded insights.
No hallucinations.
No invented trends.
Just facts distilled from the source document.
Step 2: Perplexity Finance
We downloaded the official P and L, balance sheet and cash flow statements as verified CSVs.
These came directly from filings, not scraped tables.
This gave us the clean numeric backbone needed for forecasting.
Step 3: Groq (Grok 4.1)
We fed Grok all three financial statements plus the NotebookLM summary, along with our structured prompt.
Grok handled the synthesis: the report, the layouts, the reasoning, the cross links and the valuation logic.
Because the inputs were verified, the output became significantly more accurate and aligned with institutional standards.
Why this workflow works
• Each tool does what it is best at
• Data extraction stays separate from analysis
• Grok only builds on verified inputs
• No hallucinated numbers
• No mismatched totals
• Clean, auditable outputs every time
This is now our default research stack at Shikshan Nivesh.
Why This Edition Matters
Both workflows compress days of work into minutes:
Research report drafting
Competitive mapping
Valuation layout
Tone and language analysis
Management signals
Early green shoots
Red flags
These are tasks analysts repeat every quarter.
Now they are automated, structured and exportable.
The goal is not to replace judgment.
The goal is to accelerate discovery.
Access the Prompts
Below are the exact prompts used.