Eight kinds of company we turn down. And what we’d do instead.

We’d rather lose the sale than earn a churn.

The eight, and what we tell them.

1. 01

   You’re an early-stage company.

   The engine runs on what you have already lived through: the case that went sideways, the client you talked out of a bad decision. You do not have that yet. And we map a year of content to a market you may still leave. Publish it, then pivot, and you bought a map to somewhere you don’t live.

   Do this instead Write it yourself, in public, under your own name. Today you have more time than you have hard-won answers. That reverses. Come back when it does.

2. 02

   You sell software, not judgment.

   Utopica works when your buyer is choosing a person to trust, not a product to run. Software buyers read your docs and your reviews, then start a trial. A partner’s essay is not in that path. This is about what you sell, not what you know: an MSP, a vCISO and a security consultancy all pass. The company selling the software does not.

   Do this instead Put it into documentation, a changelog people actually read, and your customers’ own words. That’s your version of this.

3. 03

   You’re a publisher.

   For our clients, the content sells something else: their expertise, their services, their name in the AI answer. A publisher inverts that. The content is the product and raw traffic is what pays. We optimize for clients, not pageviews.

   Do this instead That’s an editorial craft with its own tools: newsroom CMSes, audience teams, ad ops. Real work, different work. If your content exists to sell your expertise rather than ad space, you’re not a publisher, and this rule isn’t about you.

4. 04

   You want it on autopilot.

   We take away the writing, the research, the images, the search work and the publishing. What is left is five minutes, a few times a week. Everything that wins in AI search runs on that same input, because Google’s own advice is to showcase your internal experts. We got it to five minutes. Nobody gets it to zero, because the zero-input version at volume is what Google’s spam policy names scaled content abuse. Google Search Central, “Spam policies for Google web search”.

   Do this instead The AI blog generators sell exactly that: content with no input at all. It’s cheap, it’s fast, and it’s the content AI has learned to skip. Utopica vs AI blog generators →

5. 05

   You expect page one by Friday.

   Authority compounds over months. Anyone promising a fast climb is selling paid links, pages generated at scale, or a domain somebody else built. Google names all three as spam: sites that violate its policies “may rank lower in results or not appear in results at all.” The boost comes first. The penalty lands later, on the domain you keep. Google Search Central, “Spam policies for Google web search”.

   Do this instead If you need pipeline this quarter, buy ads. They stop the day you stop paying, and they put nothing at risk. Build the asset alongside them, not instead of them. Content marketing vs paid advertising →

6. 06

   Nobody will put their name on it.

   Every article publishes under a real person’s name. AI cites firsthand expertise, and expertise belongs to people. A company that wants an anonymous brand voice is asking for the exact content AI has learned to skip.

   Do this instead If the name is the blocker, the fix isn’t a vendor. It’s deciding which of your people is willing to be known for what they know.

7. 07

   Nobody can approve it.

   Regulated is fine. You keep review on, and every article is yours to check and sign off before it ships. What doesn’t work is a company where no single person can say yes. If every article waits for a monthly committee, the cadence dies. You don’t need fast approval. You need one person allowed to give it.

   Do this instead Settle who that person is before you talk to any content vendor, us included. Every one of them hits this same wall.

8. 08

   You’re an agency.

   We work directly with the company that holds the expertise. The month we spend building the engine is a month of learning specific people’s voice, and that does not survive being relayed. What you need is one tool your team can run across many clients. Ours is a month of custom work per company. Opposite shape.

   Do this instead Buy a platform your team can operate at your margins. There are good ones. We aren’t one of them.