# Field Notes No. 8 - Can You Scale A Hallway Conversation? By Jeff Stowell · 2026-07-30 · Insights > The Full Ratchet Society... my attempt to do some mentoring at scale. _For the next month or so, I'm working on building out "The Full Ratchet Society." This is my attempt to see if we can scale mentoring for college students in the Midwest interested in Startups, Entrepreneurship, Venture Capital, Private Equity and the like._ ## Something I'm Building For those of you joining late: I'm trying to start a mentoring effort called **"The Full Ratchet Society."** It's office hours; a recurring Zoom room for college students in Kansas, Missouri, Iowa, and Oklahoma (or students from those states who went off somewhere else for school) who are curious about venture, startups, and how capital actually works when nobody is performing for a pitch competition or trying to get a job. I've spent this last week or so trying to figure out the unglamorous parts. I've built some small administrative tools, because I know the version of this where I spend more time managing a calendar than talking to anybody. And I started doing a bit of "recruiting," which so far means emailing people I know and asking them to point one or two curious students this direction. I'm targeting the first meeting of The Full Ratchet Society for the evening of Sept 9, 2026. So if you're a young person reading this... circle that date. If you know a student who should be in the room, forward this to them. ## Something I'm Thinking About I think the reason UVF-Crossroads worked wasn't really the syllabus. It was fine. Maybe even pretty good. We taught all the stuff you need to know like term sheets and deal docs and SAFEs and cap tables and options, etc. etc. What I think worked was the twenty minutes after class when students would stick around and follow-up, or the bunny-trails we would go down when discussing new funding rounds or controversial topics. The chances when students got to ask the question they were too embarrassed to ask in front of everyone. Intuitively we know that's where a lot of teaching happens, and it happens one student at a time. In tech, that often makes for a lousy business model. It makes for even worse public good I suppose. One of the things I'm fond of saying is not **_EVERYTHING_** needs to scale. In fact, many things are better when they don't. I suspect that includes mentoring. However, we need to mentor more students in startups, venture, PE and private markets if we want ecosystems "in the middle" to grow and thrive and we have limited people and limited time to do that kind of mentoring. So how do we square that circle? I did a little reading, and education research has been staring at this problem for _at least_ forty years. I found [Benjamin Bloom's 1984 paper](http://web.mit.edu/5.95/readings/bloom-two-sigma.pdf) that demonstrated that students tutored one-to-one performed about two standard deviations better than students in a conventional classroom. For the math-challenged like me, that means the average tutored kid outperformed 98% of the control group. That's unsurprising, but how does one solve for that problem? More recent work suggests the loss isn't small either. [Kraft, Schueler and Falken](https://edworkingpapers.com/ai24-1031) re-ran the meta-analysis on tutoring in 2024 across a couple hundred randomized trials, and found that once you screen for the programs that actually look like large-scale implementations, effect sizes drop from around .40 standard deviations to somewhere in the .16 to .22 range. The benefits shrink as you grow. That is roughly the opposite of how we've been trained to think about scale. So here I sit, wondering what the part of mentoring that matters actually is. If it's a set of conditions -- a small room, real questions, an adult who's done the thing before, permission to be ignorant -- then I can build that, and it works with whoever shows up. If it's the relationship itself, accumulated over time between two specific people, then "scaling my mentoring" is a category error and I should accept a small number and do it well. My working hypothesis is, to paraphrase a mentor of mine (thanks Stephen Fawcett!), that I can create the conditions under which many students can be successful and I can maintain and foster those conditions to a point. It's not scale in the traditional sense that tech means it, but it is a "one-to-many" frame for mentoring that I think can be pulled off. ### **Some Reading** \- **Bloom, "The 2 Sigma Problem" (1984, _Educational Researcher_).** Short, readable, and the framing is better than most of what's been written since. Worth reading in the original rather than in somebody's blog summary. Linked above \- **Kraft, Schueler & Falken, "What Impacts Should We Expect from Tutoring at Scale?"** The scale-up penalty, quantified. Sobering if you're building anything mentorship-shaped. Linked above. \- **Chatterji, Delecourt, Hasan & Koning, "When Does Advice Impact Startup Performance?" (\_Strategic Management Journal\_, 2019).** A randomized field experiment with 100 Indian tech founders. Founders who got advice from peers with an active people-management style grew 28% larger and were 10 points less likely to fail. My favorite finding is the one I didn't see in many citations: founders with MBAs or accelerator experience didn't respond to the intervention at all. Prior training seems to have inoculated them against peer advice, which is either very funny or very sad depending on the day. [https://www.nber.org/papers/w24789](https://www.nber.org/papers/w24789) If you've built something that made one-to-one help go further without hollowing it out, I would love to hear about it. Thanks for reading. /JLS --- Canonical URL: https://hayseedcapital.com/blog/field-notes-8 Site: The Hayseed (https://hayseedcapital.com) — writing, coaching, teaching, and early-stage investing by Jeff Stowell. Operated by Hayseed Capital, LLC, Park City, Utah.