The Spiral Method for Learning Fast
Chain expert-to-expert referrals until you hear only repeats — that's how you know you've hit the bullseye.
- Difficulty
- Easy
- Time to result
- ~weeks to results
- Steps
- 3
- Confidence
- 92%
A method for learning a complex domain quickly AND knowing when you've learned enough. You talk to one person, understand little, then ask who else you should speak with; you repeat, spiraling inward. Early conversations are mostly new information; later ones shrink to 10%, 5%, 0% new. When you keep hearing the same things from multiple people, you're at the bullseye for your needs. It doubles as a signal-of-sufficiency, not just an acquisition tactic.
Origin
Eilon Reshef developed and wrote about the 'spiral method' (a Medium/blog post), using it to learn deep learning during a sabbatical. He likens it to annealing — material crystallizing as temperature slowly drops.
Core principles
- 01Learning has a natural stopping point: convergence, when new information drops toward zero
- 02Each expert's best gift is the next names to talk to, not just their own knowledge
- 03Tech has a helpful ecosystem — people help if you don't ask too much of any one person
- 04Aim for the level of mastery your role requires, not absolute expertise
How to run it
- 1
Talk to one person and ask for referrals
Find someone near you who knows the topic, ask them to explain it (you'll understand little), then ask the key question: who else should I be speaking with? Collect ~three names.
Pro tip The referral ask is the engine of the spiral — always leave each conversation with the next set of names.
Watch out Don't ask too much of any single person; the ecosystem stays helpful because each ask is light.
- 2
Spiral through successive conversations
Talk to the referred people, each time understanding more — the fifth person you might understand 50%, with 50% new. Keep collecting names and going inward.
Pro tip Track roughly what fraction of each conversation is new to gauge your position on the spiral.
- 3
Stop when you hit convergence
When new material drops to ~10%, 5%, or 0% and you keep hearing the same things from multiple people, you're at the bullseye for your needed level. Stop, or run another deeper spiral (user research, going all-in) if the role demands it.
Pro tip You need not become a true specialist — reaching the point where peers give you nothing new at your desired level is enough to make decisions.
In the wild
Bored on sabbatical, Reshef used the spiral method to learn deep learning: starting from near-zero understanding and spiraling through referred experts until conversations stopped yielding anything new at his desired level. He came out able to reason about ML as a product manager (and bought Nvidia stock).
→ Reached functional mastery for a PM's needs quickly, without becoming a specialist data scientist.
To serve account managers better, Reshef spiraled through conversations with account managers until their concerns converged and clearly differed from net-new sellers or contact-center sellers.
→ Reached the point of being able to make product decisions for that persona at the conversation level.
Common mistakes
Learning from a single source
Hearing something from one person means you've learned nothing reliable yet — you can't know it's representative until multiple independent people repeat it.
Is it for you?
Best for
Product leaders and generalists who need working mastery of a new domain or persona quickly and want a clear stopping signal
Not ideal for
Domains requiring certified deep expertise, or topics with no accessible community of practitioners to refer you onward
From the transcript
“you go find the person next to you and you're like what is deep learning they tell you something”
“the next question you should ask like who else should they be speaking with they give you three other names”
“I call it a spir because it's kind of going in circles around the Target”
“well if I heard it from three people I didn't learn anything new I'm sort of at the bullseye”
From the episode
Inside Gong: How teams work with design partners, their pod structure, autonomy, trust, and more
Eilon Reshef (co-founder and CPO)