--- type: WebPage name: "If you want your Selfe done right, do it with us" id: "https://selfe.ai/selfe-made" url: "https://selfe.ai/selfe-made" publisher: "did:web:selfe.ai" description: "Twelve weeks with us and an agent can buy from you. Ten businesses, cohort one starts 28 September, and we do the work." --- # If you want your Selfe done right, do it with us > Twelve weeks with us and an agent can buy from you. Ten businesses, cohort one starts 28 September, and we do the work. Agents are already shopping on behalf of your customers, and most businesses have nothing they can buy. Twelve weeks with us and you are one that they can, running it yourself by Christmas. ## Everyone running the same defaults arrives at the same answer. Getting found is the easy half. Most businesses can be made legible to an agent inside a fortnight and yours probably can too. What takes a quarter is the part after that, which is deciding what your Selfe may say, who it should turn away, what it can agree to without asking, and then watching it do all of it against real customers. Nobody arrives knowing how to do this. The job is a few months old and there is nowhere it could have been learnt. So for twelve weeks we do it with you, on your catalogue, with your money going through it, and you come out running it. Your own judgement is the only input your competitors cannot copy. Everything else is a default they have too. ## Six weeks building. Six weeks trading. The dates are not an accident. You are live by 8 November, which puts weeks seven to twelve through Black Friday and the whole Christmas run. The busiest trading of your year is also the best evidence you will ever get. ## £3,500 for cohort one. £5,000 after that. Cohort one is cheaper because you are giving us something back, which is a named case study with the week one and week twelve numbers in it. That is a trade rather than a discount, and the price goes up for cohort two either way. **Tractors & Cream** A Somerset farm café and glamping stay, connecting through ResNexus, the same adapter already proven end to end, live, on a real partner sandbox. **Memento Park** A Budapest open-air museum, connecting through FareHarbor, the same booking engine already proven live elsewhere. ## Ten places. Four questions. Every application gets read and answered either way, usually inside a week. There is no call to sit through first. What you run. How many places, or how big a range, and which systems hold the money and the calendar or the stock. Say whether your Selfe is already running, because most people apply after it is. What the defaults would get wrong about you. The one we actually read. Some way you sell that a standard playbook would flatten, whether that is a rule about who a place suits, an exception that costs you money every time somebody new gets it wrong, or a thing you always say and have never written down. What you want judging on in December. Two or three use cases. We agree them in week three and measure them in week twelve, so this is the beginning of that conversation rather than a wish. Who is in the room from your side. One person who can decide, and whoever else has to agree. Twelve weeks is long enough to settle an argument about what good looks like and short enough that it gets settled. ## Before you apply. **Twelve weeks over our busiest quarter. Is that mad?** It is the reason for the dates. You are live by 8 November, before peak starts, and the work in weeks one to six is mostly ours. What peak gives you is three months of real agent traffic instead of a quiet test. A January cohort finishes in April with nothing much to show. **Is this a course?** No. We do the work on your catalogue and your systems, and you learn by watching it happen and making the calls we cannot make for you. There is no curriculum and no homework, and if you want somebody to hand you slides then this is the wrong thing to buy. **Is this an agency arrangement?** No, and the test is which direction it points. An agency sells you more hours next quarter. This sells you fewer, because week thirteen is a subscription and the whole twelve weeks is about you leaving able to run it. The scope is capped in figures on this page rather than in a promise. **What if we are already running a Selfe?** Most applicants are, and it makes the twelve weeks better rather than redundant. A Selfe running on the standard playbooks is a considerably better thing to spend a quarter improving than a blank page. **How do we know the standard playbooks would not do?** You can usually name the thing straight away, which is the rule everybody in the building knows and nobody has written down, or the exception that costs you money every time somebody new gets it wrong. If nothing comes to mind then the defaults will very likely be fine, and the self serve route is the honest answer. **What if we do not get a place?** Ten places, and anybody we cannot fit goes to the top of the list for the next cohort at the price of the cohort they join. The self serve route stays open the whole time and has you running this week. ## Act on this - [Check whether an agent can buy from you](https://selfe.ai/agentic-commerce/buyability-check) — the free readiness scan. - [Discover venues](https://selfe.ai/api/registry/discover) — `POST`, semantic browse across the registry. - [Match a bookable answer](https://selfe.ai/api/registry/match) — `POST` with dates and party size. - [Verify Selfe's identity](https://selfe.ai/.well-known/did.json) — `did:web:selfe.ai`. - [Agent card](https://selfe.ai/.well-known/agent-card.json) — how to connect over A2A or MCP.