Core-Mystery Research Journey
Start with an unresolved question and follow people and evidence to an answer
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 98%
Hari describes beginning each book with a personal and social mystery he genuinely cannot answer. He then identifies people who may hold part of the explanation, meets them in context, and keeps asking each source who else he should see. The process expands through referrals, repeat visits, transcripts, and many interviews that produce no usable material. Only after building a large reservoir does he index scenes, statements, and themes and begin assembling them like jigsaw pieces. The method trades speed for discovery: its purpose is not to decorate a prior thesis, but to let evidence and relationships change what the writer believes before the final narrative is built.
Origin
Hari explains on The Diary of a CEO how he researches books such as Lost Connections and Chasing the Scream through unresolved questions, extensive travel, repeated interviews, referrals, transcripts, and long periods of synthesis.
Core principles
- 01Begin with a question whose answer you do not already know
- 02Find other people rather than trying to find yourself
- 03Trust accumulates through repeated contact
- 04Dead ends are part of discovery
- 05Assemble an explanation only after gathering enough pieces
How to run it
- 1
Name the mystery
Write the central question in a form that matters personally and socially. Confirm that you do not already know the answer.
Pro tip Use one question strong enough to organize the whole inquiry.
Watch out Do not disguise a fixed conclusion as an open question.
- 2
Enter the real setting
Find people with relevant expertise or lived experience and meet them where the subject is unfolding when practical.
Pro tip In-person contact can reveal context and build trust that a screen misses.
Watch out One vivid source is not enough to settle a broad claim.
- 3
Follow the referral chain
End interviews by asking who else you should meet. Follow successive leads until you reach sources who materially change the inquiry.
Pro tip Treat each useful source as a pointer, not the finish line.
- 4
Build the reservoir
Record, transcribe, revisit, and index the material. Accept that many interviews and pages will not appear in the final work.
Pro tip Mark scenes, claims, contradictions, and recurring themes separately.
Watch out Low-yield interviews are a normal cost, not proof that the search has failed.
- 5
Assemble the explanation
Group related pieces and construct the narrative only after the evidence base is broad enough to challenge your starting assumptions.
Pro tip Let the strongest material determine the structure.
Watch out Do not force every collected piece into the final answer.
In the wild
Hari says he interviewed many people about childhood trauma, depression, and addiction before a referral chain of roughly five people led him to physician Vincent Felitti. Felitti's account of the obesity program and later adverse-childhood-experiences study became a major explanatory piece in Hari's work.
→ A long referral chain surfaced a source Hari considered central to understanding the subject.
A reporter starts with the open question of why users abandon a service. Interviews with customers lead to support agents, former employees, and a researcher. The reporter indexes repeated explanations, contradictory cases, and concrete scenes before deciding what mechanism the evidence supports.
→ The final account reflects discovered causes rather than the reporter's first guess.
Common mistakes
Starting with the answer
Hari says he cannot imagine writing these books from a position of already knowing what he will teach the reader.
Rejecting dead ends
The search requires tolerance for interviews and material that contribute nothing directly to the finished work.
Expecting instant disclosure
Hari says people often open up only after sustained contact rather than during the first meeting.
Is it for you?
Best for
Writers, journalists, and investigators exploring a complex question through evidence and human experience.
Not ideal for
Time-sensitive reporting that requires a reliable answer within hours or days.
From the transcript
“I always start with a core question.”
“At the end of every interview you say, who else should I talk to?”
“You have to have a high tolerance for dead ends.”
From the episode
Johann Hari - Everything You Think You Know About Meaning & Happiness Is Wrong