Every report a consultant reads. None of the paywalls.
The research sources I actually use for case competitions, internships and decks. MBB insight hubs, India's official data, global institutions and the best free academic libraries, sorted by what you're trying to find.
Curated by Shivam Meethelping you win at everything
120K+ on Instagram
hand-picked
to India data
no login at all
you need to buy
Research like a consultant.
Good research isn't reading more. It's reading in the right order. This is the loop I run for every case, from a 24-hour sprint to a national final.
Frame the question
Write the one question your deck must answer before you open a single tab.
Size it with hard data
Start with official numbers. Government and institutional data is the most citable thing you can put on a slide.
Borrow a point of view
Consulting reports tell you what the smart money thinks is changing, and which frameworks they use.
Triangulate
Never trust one number. Find two independent sources and explain the gap between them.
Cite it properly
Source, year, and page on every chart. Judges notice, and it makes your numbers bulletproof in Q&A.
What are you researching?
Skip the scrolling. Pick the job you're doing and I'll filter the library down to the sources I'd open first.
54 sources. Zero fluff.
Filter by category or by how open a source is. Press / to search. Tap the bookmark on anything you'll come back to and it saves to your stack on this device.
Access status checked when curated. Big sites sometimes move pages, so if a link breaks, search the source name.
How I turn reading into a deck
Read like a consultant.
Think out loud.
Finding the report is the easy part. The hours go into what comes after: notes, synthesis, AI prompts, slide copy. So I stopped typing that part. I speak it into Wispr Flow, and it lands clean in Notion, Docs or ChatGPT. Here's why that's not just a habit, it's backed by research.
Speaking beats typing, on speed and accuracy.
In a study of 32 people on an iPhone, English speech input was about 3 times faster than the keyboard, with a 20.4% lower error rate. That was 2016 speech recognition. It has only improved.
Explaining while you read builds real understanding.
Chi and colleagues had students explain each line of a text to themselves as they read. They learned more than students who simply read it twice, and the students who explained the most all reached the correct mental model.
Spoken prompts get better AI answers.
When you type a prompt, you cut it short. When you talk, you naturally include your case, your constraint and what you already know. That context is exactly what turns a generic ChatGPT or Claude answer into one you can actually use.
I love thinking out loud, and Wispr Flow turned it into my workflow. It turns my “okay, hear me out” moments into scripts, emails, and AI prompts. Less time typing, more time doing :)
~3,000 words of notes, prompts and slide copy at ~40 words a minute.
Same 3,000 words at ~130 words a minute.
Almost an hour back on every case. That's two more reports read properly, or one more round of slide polish.
Talk to the report
Open the PDF next to ChatGPT or Claude and ask questions out loud. Long, specific prompts cost you nothing to say.
Margin notes by voice
Hold the key, say what the chart means, let go. Clean bullets land in Notion or Docs while the report stays open.
Jargon spelled right
Add CAGR, EBITDA, MOSPI, NITI Aayog and your case company to the dictionary once. Never fix them again.
Citation snippets
Save your source-line format as a snippet. Say the short cue and the full citation drops in, every time.
Edit by voice
Highlight a messy paragraph and say "turn this into three bullets". That's Command Mode, and it's how my notes become slide copy.
Everywhere you research
Docs, Notion, Slides, Sheets, Gmail, any chat box. On Mac, Windows, iPhone and Android.
Get 3 months of Wispr Flow Pro, free.
My code SHIVAMEETSWISPR is applied automatically through this link.
Time estimates assume ~40 wpm typing and ~130 wpm speaking; your speeds will vary. Research: Stanford News, 2016 · Chi et al., Cognitive Science 18(3), 1994.
6 research mistakes judges catch.
I've judged and competed in enough case rounds to see the same slips again and again. Every one of these costs you in Q&A.
One source for a market size
A single number with no cross-check is the first thing a judge pokes at.
Stale numbers
A five-year-old stat in a fast market quietly kills your credibility.
Forecasts presented as facts
A consulting projection is an estimate, not a measurement.
No source on the slide
Uncited numbers look made up, even when they aren't.
Citing what you can't open
Quoting a paywalled Statista or Gartner chart you never read is risky.
Ignoring government data
MOSPI, RBI and NITI Aayog are free, authoritative and underused.
@meet_.shivam
Hi, I'm Shivam. I'm helping you win at everything.
This library is every source I kept coming back to across 25+ case competition wins and a stack of internships. I put it in one place so you can spend your hours thinking, not hunting for a PDF.
on Instagram
wins
(Unstoppable Leader)
The research is free.
The edge is what you do with it.
Pick one question. Open two sources. Say what you learned out loud. Do that for every slide and you'll out-research most of the room.