Sample report · Wave One
This is an abridged version of what a wave produces, published so you can judge the work before you buy any of it. The findings are real and the numbers are unmodified. What is missing is the part clients pay for: verbatims, the raw anonymised set, and the cross-tabs by profession.
Self-completion instrument, fielded to the Grateful Giraffes membership over six days with no incentive offered. 58 completes, no partials. Multi-select answers are counted by matching whole option strings, never by splitting stored text on punctuation, since several option labels contain commas and splitting them inflates every share. Where a question is a ladder, we report the furthest rung a respondent selected rather than counting them at every rung. Multi-select shares therefore sum past 100%. Subgroups below 30 are reported as direction, not as precision, and are labelled as such.
The population is a membership community, not a probability sample of any wider market. It is concentrated in Los Angeles. Read it as a deep account of a specific, hard-to-reach population rather than as a national estimate, because that is what it is.
Adoption does not taper the way a maturity curve says it should. Four in ten are still circling and three in ten already ship. Anything built for the middle of this distribution serves neither end of it, which is the single most expensive mistake a product can make here.
n=58. Furthest stage each respondent selected; someone who picked both Curious and Building is counted once, at Building.
This is the constraint everything else has to fit inside. Any offer, product or curriculum that assumes more than five hours of weekly attention is designed for 15% of this population. The intent is real; the available time is not.
n=58. Highest hours band each respondent selected, so this is a ceiling and not an average. Reported as ceilings because 17 respondents selected more than one band.
Agents and operations run away with it. A wellness and creative community ranking creative AI last, at less than half the interest in agents, is exactly backwards from what the room looks like — and it is the kind of thing you only ever learn by asking rather than assuming.
n=43–50 per topic. Share rating each topic 8 or higher on a 1–10 scale. Topic means range from 8.76 (agents) to 5.78 (creative).
Beginners ask for a teacher and a hand to hold. The people already shipping want to be thrown in a room for a weekend and left there. Only the small-group core holds across both. A topline would have shown none of this — it appears solely when answers are crossed against who gave them, which is the entire argument for a panel with profiles behind it.
Early = Curious + Dabbling (n=24). Late = Shipping + Teaching (n=18). Multi-select, shares sum past 100. Small subgroups: direction, not precision.
Only one in six wants this to stay virtual. For a population this far into the technology, the pull toward physical presence is the counter-intuitive result of the wave, and it is the one with the clearest commercial consequence for anyone selling to them.
n=58. Share selecting in-person only or the in-person-and-virtual option, against those selecting virtual options exclusively.
Four things, deliberately. Three of them are the deliverable. One of them is a promise we made to our members.
Every respondent wrote, in their own words, what they are trying to achieve in the next thirty days. 58 distinct answers, no two alike. They are the most useful thing in the dataset and they are shared under NDA, never publicly, because our members wrote them for us and not for the internet.
Every field, every respondent, ready to cut yourself.
The same questions crossed against 526 stated professional backgrounds. Practitioners answer differently from founders, and that gap is usually where the commercial decision lives.
Waves can field as recorded, transcribed and diarised conversations rather than as checkboxes. Text tells you what someone picked. Voice tells you where they hesitated before picking it.
A first wave is $7,500 and the full fee is credited against anything annual you do next. If we do not think this population can answer your question, we will say so on the call.