graham duncan — talent is the best asset class

the core question

"what's going on here, with this human?"

this is the fundamental question graham asks when evaluating talent. not "are they smart?" or "do they have the right credentials?" but a genuine curiosity about the whole person.

finding people for positive feedback loops

graham's mission: get as many people as possible into positive feedback loops. the key insight is that you don't want to put something into somebody — you want them to do the thing they want to do anyway.

the robert story: graham was interviewing a candidate for an analyst role. the candidate's eyes didn't light up when discussing analysis. but then the candidate stalked graham for four weeks — calling at home, emailing twice a day. revelation: this person should be in sales, not analysis. by putting robert in the right seat, he built their entire business.

the rider and the elephant

borrowed from jonathan haidt's metaphor: the conscious mind (rider) thinks it's in charge, but the unconscious (elephant) actually drives most behavior.

• the rider = conscious mind, verbal reasoning, what people say they want
• the elephant = unconscious drives, what actually motivates them

in interviews, most people present their rider. the challenge is seeing the elephant — what really drives them, what they're compulsive about, what they can't help doing.

questions that reveal the elephant

• "what are you compulsive about?"
• "where have you experienced a moment of ignition — when you saw an older person doing something and intuited they were wired like you?"
• "how would your spouse/sibling/parent describe you with ten adjectives?"

the ten adjectives question works roughly 50% of the time. candidates capable of taking someone else's perspective (more common in older candidates) reveal both rider and elephant.

the power of stillness in interviews

the easiest way to separate signal from noise: talk very little. create stillness.

another technique: have the candidate ask you questions. questions have very high signal value. graham writes down each question and sometimes responds with "i'll answer, but first — why did you ask that?"

he's looking for the felt sense of a "hungry mind" based on how questions flow. that's hard to fake.

references over interviews

"i now consider in-person references with someone who knows the candidate well 5x more valuable than an interview. they can be 10x more valuable when you're in a high-trust relationship with the reference-giver."

the table-pounding reference: graham remembers a CFO reference where the woman's tone was "why are you wasting time talking to me instead of convincing him to join you?" she felt sorry he didn't get it yet. that tone is the highest signal.

the dog that doesn't bark: if you don't hear table-pounding from someone who knows the candidate well, is it contextual (they're tired, jealous, etc.) or is there real signal? this requires 10,000 hours to calibrate.

resourcefulness as meta-competency

graham cites brad stulberg: resourcefulness is the single most important competency.

"imagine you have magical glasses that flash green when a candidate reveals resourcefulness and red when they don't. you must constantly ask yourself: does this example show resourcefulness or lack of it?"

context and ecosystem

a portfolio manager might run a large idea by their boss, and subtle reactions (raised eyebrow, long pause) cause them to size the position smaller. that tension is constructive. remove them from that container and they won't perform the same way.

this especially applies to people starting their own companies — they may struggle to recreate the culture that allowed them to thrive. strengths in one context can become weaknesses in another.

everyone is an A player at something

"i try to stick to the default assumption that everyone is an A player at something."

it's more effective to approach interviews as a live puzzle — discovering what elephant and rider do well — rather than binary judgment about whether someone is an A player.

key insight: genius next to dysfunction

one marker that you're seeing someone clearly: when you understand how their strengths are also their weaknesses, how their genius lives right next to their dysfunction.

being overly excited or overly skeptical means you're further from clarity.

taste vs judgment

graham prefers "taste" over "judgment" — judgment implies one right answer, whereas taste acknowledges personal preference and pattern recognition. you develop taste in people over time, and it evolves.