2026-10-28 ยท 5 min read # Known, Trusted, Loved, Chosen The product, the proof, the language and the service all shape whether someone chooses us. AI is making brand more valuable. It has made intelligence cheap and software close to free to create. What cheap intelligence finds hardest to reproduce is experience, wisdom, and knowing what to build, for whom, and why it matters to them. Companies that understand their customers that well are the ones people trust and love. As features converge, brand is demand. The product, the proof, the language and the service all shape whether someone chooses us. They need to work together, because to the person considering us, it's all the same company. I've been describing the work with four words: known, trusted, loved, chosen. They're my version of the tried and true awareness, consideration, conversion, loyalty funnel, modified for the AI era. ## Known People need to know we exist before they can consider us. They need to know what we do and why it matters to them. It's easier to generate a lot of content. Everyone has turned the noise up to 11, which hasn't made getting noticed any easier. Baseten felt bigger to me because I kept running into them: their website, an ad on Reddit I accidentally clicked, then X. People already know what an established company does, and that understanding can be hard to change when the product changes. If we keep telling the old story, we reinforce the old perception. Buyers' agents need to find us too. A buyer's first encounter with us may be an AI answer. At Blaxel, I built a tool that turned answer-engine citation data into a weekly queue of Reddit conversations to review. I describe that work in [Be the Reference](/writing/agentic-traffic-aeo-2026). ## Trusted I've spent some time working out whether trusted or loved comes next. I'm settling pretty strongly on trusted. When someone tries our product, they need to see its value quickly. People lose trust when it breaks or doesn't do what was promised. Dave Steer describes marketers as [the trust engineers of go-to-market](https://www.gotomarketalliance.com/trust-is-the-differentiator-in-your-go-to-market/). Trust equals consistency over time. We make a promise in the language we use, the proof we show and the experience we offer. The product has to deliver on it, and people need to know we'll show up when it doesn't. At Blaxel, I paired customer stories with GitHub cookbook examples. Buyers and their agents can inspect the code. People also need to know what happens when something goes wrong. When I ran my market signal agent on Spur, in AI software validation, one commenter wanted to see the screenshot before an agent changed the selectors the test used to find things on the page. The category was leading with the opposite demo: first test in five minutes. The walkthrough I wanted was a checkout run that failed: show the evidence, the diagnosis, and the review step before the tool changes anything. My read of the signal was that the category was trying to jump from known to chosen with a five-minute demo. The market is indexing toward trusted. Trust can be marketed only after it's earned. Communicating technically enough to be trusted, but simply enough to be understood, sums up my job in every role I've had. ## Loved I do mean the brand should feel cool. I want the product to be nice to use, and I want people to like being associated with the company. In a conversation recently, I used the example of a Prime Intellect hoodie. Someone sees you wearing it and says, "Where'd you get that? That's cool." That's part of what I mean by loved. There's also the feeling that a company understands your world. It speaks your language, the product fits your work, and when you need help, someone cares about solving the problem with you. Technical audiences value brand. For them, it's mostly language, positioning, and how and where the company shows up, with a little logo and colour. I can be drawn to a brand before I've used it enough to trust it, and I can trust a product without particularly loving it. Trust and love work together, and both contribute to that first choice. ## Chosen When someone is ready to try or buy, do they pick us? The first choice matters most. We have to earn it before we can become someone's default. I want people to choose us again because of the value they get, rather than the work it would take to leave. That's the argument in [The Lock-In Is Moving](/writing/the-lock-in-is-moving). A customer story helps someone understand what they could do with the product, and a working example gives them somewhere to start. Their experience with the product changes how they hear the next thing we say. Those things change how the others land. When we get them working together, we create a feeling about the company. When people aren't choosing us, we need to know why. Do the right people know what we do? Do they understand how the product fits their work and believe it'll do what we promise? Did trying it give them a reason to choose us again? We need to answer those questions before we commit to a campaign, a new website or a hire. ## When the agent chooses For an enterprise deal, an agent can do the first pass of research. The person taking that recommendation to a buying committee or CFO still needs confidence in the company. Someone commenting on my first post about the four words asked whether loved and chosen should swap places as software becomes cheaper and easier to implement. I can see it: a product could be known and already trusted, then chosen by the agent. Loved still belongs. What I haven't worked out is what trusted and loved mean when the one choosing is a machine. ## Questions this piece answers ### What do known, trusted, loved and chosen mean? People need to know we exist and understand why the product matters to them. Trust comes from delivering on our promise consistently. Loved includes liking the product, feeling understood and wanting to be associated with the company. Chosen asks whether people pick us when they are ready to try or buy. ### Does trusted always come before loved? I'm settling pretty strongly on trusted, but I can be drawn to a brand before I've used it enough to trust it. I can also trust a product without particularly loving it. Trust and love work together, and both contribute to that first choice. ### What changes when an agent chooses the product? An agent can do the first pass of research while the person taking its recommendation to a buying committee still needs confidence in the company. A product could also be known and already trusted, then chosen by the agent. What I haven't worked out is what trusted and loved mean when the one choosing is a machine.