--- title: "Curation" description: "When supply exceeds anyone's ability to evaluate it, the edge goes to whoever evaluates for others and says no most of the time." canonical: https://moa.giglabo.com/strategies/curation/ locale: en --- # Curation > Markdown twin of https://moa.giglabo.com/strategies/curation/ — the same content as the HTML page, a fraction of the bytes. > Every page of this site has one: append `.md` to any URL, or `index.md` to a directory URL. > Site structure and the full page list for agents: https://moa.giglabo.com/llms.txt Category: Accreditation · Role: Holds · Depth: 4 Gist: selectiveness in choosing and grouping ## Moats - [#33 Category leadership](https://moa.giglabo.com/moats/33/) (direct, depth 1 · Shovel) — A visible, consistent set of choices in one niche makes your name the shorthand for that category fast; it is the quickest moat here and the shallowest, since a rival with a comparable list catches up in a season. - [#8 Brand and trust](https://moa.giglabo.com/moats/8/) (direct, depth 4 · Mine) — Years of picks that hold up turn the selector's name into the guarantee; the buyer trusts the label instead of testing the item, and that trust takes a rival a decade to match. ## Summary When supply exceeds anyone's ability to evaluate it, the edge goes to whoever evaluates for others and says no most of the time. The value is in what is left out: the buyer trusts the selection because the selector rejects. Grouping adds a second layer — a roster, a slate, a portfolio becomes a name people look for on its own. ## How it works The mechanism is rejection with a track record. Each pick that holds up raises the price of the next one, because the buyer stops evaluating the item and starts evaluating the selector — once. A selector who accepts everything is a directory; a selector who rejects most of what qualifies is a filter, and the filter is what the buyer pays for. Grouping compounds it: items placed together take on a meaning none of them had alone, and the group gets a name — a gallery's roster, a network's slate, a fund's portfolio. The supply side then comes to you. Being chosen is itself a mark, so makers, artists and founders apply to the selector whose list is shortest and best kept, not the one that pays most. That flow is the edge: you see the candidates first, and every strong one you accept makes the next rejection more credible. ## Why it holds — or only takes It holds, but starts shallow. Category leadership is a shovel, depth 1: a consistent taste in one niche makes your name the shorthand for it within months, and a rival with a comparable list takes it back within a season; it is under attack from AI, because recommendation at scale is what models do. Brand and trust is a mine, depth 4: if the picks keep paying off for years, the label becomes the guarantee and the buyer stops checking the item. The whole game is the crossing from the first to the second — every batch of picks either builds the name or spends it. ## Do you have it - People ask what you have picked before they ask what it costs, and your rejections get quoted. - Makers, artists or founders want onto your list more than onto a longer one. - Buyers take an item on your selection alone, without checking it themselves. - A rival's list is compared with yours, never the other way round. - Being dropped from your list costs the dropped party something visible. ## How to dig it - State the criterion in one sentence and reject most of what meets it; a list that grows with demand stops being a list. - Put your name on the group, not only the items: a series, a roster, a fund the buyer can subscribe to as a unit. - Keep a public record of past picks so the track record is checkable; that is what turns taste into trust. - Say no visibly, and make being chosen worth something to the chosen — promotion, a fee, a mark — so supply comes to you. - Keep the selector's taste in the house, not in one hire: document the criteria and let more than one person apply them. ## How it erodes - The rejection rate falls as you scale; the list gets long and the name stops meaning chosen. - A recommendation model hands every buyer a personal selection, cheaper than yours and tuned to them. - The curator leaves, and taste that was attached to a person walks out with them. - A pick fails in public and the label pays more for it than it earned from the wins. - The whole catalogue becomes searchable and cheap; the buyer no longer needs anyone to pre-select. ## Solo, without capital This is a solo builder's strategy: Category leadership is reachable solo and costs a shovel. One person with a clear criterion and a public track record can be the list for one niche within a quarter — a newsletter, a directory, a curated shop. What you cannot buy is the mine: trust at depth 4 is years of picks that held, and a solo curator's brand is the person, so it neither scales nor survives them. Start by publishing the picks and the rejections; the record is the asset. What the matrix says of its moats: - [#33 Category leadership](https://moa.giglabo.com/moats/33/) — Solo: ✓ solo-reachable · Capital: ◐ medium · AI: ↓ under attack · Rent: ✗ not for sale - [#8 Brand and trust](https://moa.giglabo.com/moats/8/) — Solo: ~ partly · Capital: ◐ medium · AI: ↑ AI-resistant · Rent: ~ partly ## In nature, in business Nature: — Business: art galleries, HBO, venture capital - **A24** — A24 picks a small slate of films and puts its own name in front of them, so audiences buy tickets on the label before they know the director. - **Costco** — Costco stocks a fraction of a supermarket's range and sells much of it under one house label, so the short shelf promises that the choosing was done for you. - **Criterion Collection** — The Criterion Collection releases a chosen set of films under one imprint; a Criterion edition marks a film as worth owning before anyone reads a review. ## Combines with - [Prestige](https://moa.giglabo.com/strategies/prestige/) — Prestige is what the curator hands out: being on the list is an endorsement, and the endorser's name grows with each pick that holds. - [Scarcity](https://moa.giglabo.com/strategies/scarcity/) — Scarcity is the curator's discipline: the list stays short by decision, and the shortness is what the buyer reads as quality. - [Specialization](https://moa.giglabo.com/strategies/specialization/) — Specialization gives the criterion its edge: a selector who knows one niche deeply rejects with reasons a generalist cannot see. - [Authenticity](https://moa.giglabo.com/strategies/authenticity/) — Authenticity is the second thing a curator vouches for; where fakes are common, the roster is also a provenance record. [This strategy on the map](https://moa.giglabo.com/strategies/?view=lanes&hl=curation) [Map this combination](https://moa.giglabo.com/strategies/?view=lanes&hl=curation,prestige,scarcity,specialization,authenticity) ## Fights with - [Completeness](https://moa.giglabo.com/strategies/completeness/) — Completeness wants everything on one shelf; curation earns its price by leaving most of it off. - [Aggregation](https://moa.giglabo.com/strategies/aggregation/) — Aggregation opens the door to broad participation; a selector who lets everyone in has nothing left to select. [Map this tension](https://moa.giglabo.com/strategies/?view=lanes&hl=curation,completeness,aggregation) ## Source Strategy taxonomy © kepano (Steph Ango), paraphrased with attribution. Role and moat mapping © Moat Atlas. https://stephango.com/ ## Related - HTML version of this page: https://moa.giglabo.com/strategies/curation/ - Русский: https://moa.giglabo.com/ru/strategies/curation/ - Previous: Prestige — https://moa.giglabo.com/strategies/prestige/ - Next: Monopoly — https://moa.giglabo.com/strategies/monopoly/ - Atlas: https://moa.giglabo.com/ - Catalogue: https://moa.giglabo.com/moats/ - Calculator: https://moa.giglabo.com/calculator/ - Strategies: https://moa.giglabo.com/strategies/ - Site map for agents: https://moa.giglabo.com/llms.txt