Socrates as a Service

Was this newsletter forwarded to you? Sign up to get it in your inbox.

I’m a journalist and a communications expert. My job, in both roles, is to find ideas that people haven’t yet put into words—the anecdote that could become a front-page story, the framing that could crystallize a founder’s philosophy into something a customer remembers.

In an hour interview with someone, it might not be until minute 45 that we start getting into the good stuff. In two hours, there may only be one thing that stands out to me—a side story, a detail, some color. A little piece of gold dust. An investor I’ve worked closely with calls these “extraction sessions.” I call the people who do them well Socrates-as-a-service.

Those details and stories aren’t on the internet. They’re not in any model. And the model hasn’t replicated yet how I pull them out of people. The gap between what AI can do and what a great human questioner can surface is still wide—and it’s the gap where the best stories live. If you don’t have some way to surface that information in your organization, your brand and messaging are going to sound like all the other twice-boiled content out there.

Osakan bread and the wisdom within

The stuff that I’m looking for has a name in management theory: “tacit knowledge.” The term comes from scientist and philosopher Michael Polanyi, who defined it with the phrase, “We can know more than we can tell.” It’s the expertise and intuition that lives in our bodies and resists being turned into a document.

In a frequently cited 1991 article, Japanese management expert Ikujiro Nonaka argued that while Western companies excelled at “information processing,” Japanese companies specialized in the “creation of knowledge,” through a feedback loop that turned tacit knowledge into a competitive advantage. His most memorable example: In the 1980s, the Osaka-based Matsushita Electric Company was struggling to get the kneading right in a bread machine. They sent a software developer to apprentice with a baker at a local hotel famous for its luscious loaves. The knowledge she brought back helped the team perfect the dough-stretching technology inside the machine and ultimately create a top-selling device.

I am sure that the lucky engineer asked the baker a lot of questions, but there was certainly a lot she absorbed just from watching. Indeed, Polanyi argued that tacit knowledge exists outside of numbers or symbolic language—the kind of systemization that AI requires to ingest information.

Many “bakers” from whom we try to extract tacit knowledge often don’t even know the depth of expertise they carry. And they certainly couldn’t tell you what questions you need to ask to access it.

Your CRM is programmable now. In English.

You’ve done the same pre-call routine hundreds of times: LinkedIn, Crunchbase, inbox, last transcript. Fifteen minutes, every time. Lightfield is an AI-native CRM that just shipped Skills. Describe any workflow in plain English—call prep, deal scoring, account research, pipeline forecasting—and the agent learns it. “Prep me for my call with Acme.” One sentence pulls every email, transcript, and deal note and writes your brief. The agent runs against your full CRM data graph with code execution, web search, and file I/O. More than 3,000 startups are already on the platform. Use code EVERY13 for 3 months free.

Try Lightfield for free

Want to sponsor Every? Click here.

AI as an imperfect interlocutor

AI can do some of that questioning and, in some cases, do it well. At Every, we have an AI agent ask us questions when we write OKRs. The agent has ingested Every’s company strategy and has context on all the members of the organization. My colleague, Katie Parrott, has Claude interview her before she writes an article. Those notes become the basis of an outline of the piece.

I would argue, however, that AI-driven extraction works well when the parameters are clear and the assignment structured, like writing an article or a plan for software. If you’re looking to turn over a completely new rock, interview someone about something they haven’t spoken much about before, or run the kind of open-ended information gathering work that happens when companies decide to rebrand. In those sessions, a chief marketing officer or branding agency will spend time speaking to members of the company and asking them open-ended questions about the business. The point is to keep things open, go wide, and see what comes up.

There’s a second problem: A human in the room can be surprised mid-conversation and abandon the plan—perhaps notice hesitation or dig into a thread that wasn’t on the list. A prompt mostly can’t. When I elicit insight from someone, I am applying my judgment about what is a good story in real time—judgment that’s been honed by years in news and communications. This mutual, live attention is something AI can’t capture because it’s not in the room.

The obvious objection is that this is a moving target—context windows and memory are improving to allow for more detailed, fluid conversations. Taste won’t, however, won’t. Someone still has to decide which detail out of a two-hour conversation is the piece of gold dust.

Nonaka himself argued that the goal isn’t always to make tacit knowledge fully explicit. Because tacit knowledge is so personal and often so abstract, sometimes the right tool with which to communicate is a metaphor or an analogy—a form of language that can hold multiple ambiguous meanings. Eliciting that kind of language from someone takes its own form of tacit knowledge: the skills of a Socrates.

Steal these techniques

So how can you surface those nuggets of gold? Despite the explosion of interview podcasts asking for multiple hours of your time, I find most hosts are not great at asking questions. The format demands an arc—a journey—which is the opposite of what you want when you’re trying to surface tacit knowledge. Real extraction zigs and zags, doubling back on itself and picking up something you said 20 minutes ago to pull a different thread, following gold, not audience interest.

Here are the techniques that I keep coming back to:

• Warm people up. We open up more once trust is established. I never skip the small talk at the beginning of a conversation, and I’ll often bring something we have in common: “I saw you just spoke about X—I’ve been thinking about that too.” NPR interviewer Terry Gross’s favorite icebreaker question is, “Tell me about yourself.” The question lets the person you are speaking to take the lead and protects you as the questioner from saying anything that might make them prickle while you are still warming up.
• Ask a mix of general and specific questions. When Lenny Rachitsky revealed the questions he sends to his podcast guests in preparation for the podcast, this combination stood out. For example, he asks them, “Anything you haven’t shared elsewhere that could be interesting to share in this forum?”—a very general question, and “What’s one pivotal moment in your career?”—which asks the guest to pinpoint one turning point. To extract unverbalized insights from someone, it helps to ask them to both think macro about their area of expertise as well as micro.
• Come back to thoughts and drill in. If a line of inquiry goes nowhere, don’t abandon it—go back later and try again from a different angle. The first pass often loosens wisdom up.
• Repeat things back. Repeating what someone said often helps them process their thoughts further, and they will often add additional detail they didn’t know they remembered.
• Detail, detail, detail. Specifics are where the real stuff lives. How did that make you feel in that moment? Why do you think that way?
• Listen well. Pulitzer Prize-winning radio journalist Studs Terkel spent decades interviewing everyday people in Chicago, and was described by one subject as offering “a state of being, it’s a way of attending to, attention-ing another person.” That is what good listening looks like.
• Ask about squirrels. In his documentary about the debate surrounding the death penalty, Werner Herzog interviews a death row chaplain who, at the start of their conversation, delivers the polished answers he’s given 100 times about accompanying people in their final minutes. Then Herzog asks him about squirrels. Thrown off, he breaks down. The grief he feels about his job is laid bare. Ask people about the unscripted things.

Study this. Collect great questions you like. Build prompts to borrow these techniques for structured AI-driven sessions if you want.

But the judgment underneath these habits remains harder to transfer. It’s its own form of tacit knowledge. And for now, it still belongs to humans.

Eleanor Warnock is the managing editor at Every. She has been a business journalist and editor at the Wall Street Journal and the Financial Times-backed Sifted, and is an advisor to Bek Ventures. Follow her on LinkedIn and Substack.

To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.

For sponsorship opportunities, reach out to [email protected].