James Graham.

The schooling question we could not settle over dinner.

Counsel.day grew out of a real question. Whether to move into the city for our son's schooling, or stay in the small village in the countryside we had moved to before he was born. The conversation went round the table for the better part of a year. Some evenings we agreed; some evenings we did not. The thing we kept noticing was that on Tuesday the answer was obviously yes, and on Thursday the answer was obviously no, and the difference between Tuesday and Thursday was not new information · it was us.

I work with time-series for a living. Long-running streams of measurements taken at regular intervals, each one a partial signal, the aggregate carrying information no single measurement can. The structure that solves this class of problem in my day job · capture the same observation across the period, then read the whole stream · is the structure underneath Counsel.day. It was not a leap from one domain to the other. It was the same problem, presented in the kitchen instead of in the data.

The first version of the product was a series of weekends. One sentence each evening, sealed from each other, a single readable document on day thirty. We ran it on the schooling decision. The document we read on day thirty was not the answer, exactly · but it was the most honest reading of what each of us had actually thought, across thirty separate evenings, that either of us had ever seen. We made the decision the following weekend, in a conversation that lasted less than two hours.

The same workshop that has already shipped a small handful of useful, durable products.

Counsel.day joins three siblings from the same small workshop. Each is operational on its own terms · none raised outside capital, none pivoted into a different category, none is for sale. The pattern is small instruments that solve a specific class of problem well, built and maintained over a long horizon by a small team. None of the data passing through any of these products has ever been sold, despite repeated commercial offers.

The path from the North East of England to a small village in Canterbury.

I read Human Geography and Social Anthropology at university · two disciplines that spend a lot of time taking the question apart before they try to answer it. That habit has stayed with me, and it is the habit underneath every product on the workshop's bench. The early web-design and tool-building started in the early 2000s, between lectures and after work. In 2009 the New Zealand government's Be Your Own Boss programme, run through Trade and Enterprise, awarded me a grant to keep going. The workshop has been operating in some form ever since, and at one point ranked among the top ten Google Publishers in New Zealand.

I do not appear on television or radio. I have declined every invitation. The work is the point, the work is what should be visible · not the person making it. If you have read this far, this is roughly as personal as it gets.

Useful tools are a right, not a privilege.

The free products in the workshop exist because of a single conviction: financial literacy is not a privilege reserved for the wealthy · it is a basic right that every person deserves. That sentence has been the steering rule for the workshop for a decade. The audience the workshop builds for is the person sitting at their kitchen table at 10 pm, trying to figure out whether they can afford their first home · not the institution that has already figured it out.

Counsel.day is the workshop's first paid product, and the reason it is paid is the same as the reason the others are free: the cost has to match the intent. The decisions Counsel.day handles are not commercial questions, and they cannot survive the presence of advertising. Charging a small amount per decision keeps the product free of any incentive other than the user's own. No ads, no data sold, no funnel.

What I am not.

I am not a therapist. I am not a counsellor. I am not a psychologist or clinical psychologist. I am not a medical professional of any kind, and I have no clinical training. Counsel.day has not been designed, validated, or endorsed by any clinical body, and we make no clinical claims about it. The product is built around findings from decision science (the research is cited on the therapists page), but the product itself is a small private instrument for one specific decision, not a clinical intervention.

If your question needs clinical attention · a mental-health concern, a safety concern, a relationship in crisis · please reach out to a qualified practitioner in your country. Counsel.day is not the right answer there. The product is for people working through a decision they have time to weigh, not for people in distress.

Small, durable, premium · on its own terms.

The workshop is solo most of the time, with occasional collaboration on specific pieces. Counsel.day is built and maintained by me, with my wife Alexandra contributing the original observation that started the product and ongoing feedback on the editorial and product decisions. The verdict pipeline uses our AI synthesis tool API on the verdict day to generate the synthesis · no other AI service touches your data, and no other party sees your votes or notes. The marketing copy was AI-assisted under my direct editorial supervision; the methodology, the research positioning, the boundaries, and the decision-science citations are mine. Where AI was used in this site is documented in plain language at the top of every long-form page.

I plan to be running Counsel.day in twenty years. The product is funded by the per-decision price and the annual subscription, plus revenue from the other two products in the same workshop. No venture funding. No data sold. No advertising in-app, ever. The decision a couple runs inside the product is never tracked, measured, sold, or advertised against. If any of that ever changes, you will hear it from me in plain language before it happens.