How to Make Your Product Remarkable: Run a Purple Cow Audit in 5 Steps

Most products don't have a marketing problem. They have a remarkability problem.

That's Seth Godin's argument in Purple Cow: driving past fields of brown cows, you'd stop for a purple one. "Remarkable" isn't a compliment — it's a literal test. Would a customer remark on this? Would they bring it up, unprompted, to someone else? If not, more ad spend just shows a brown cow to more people.

Reading that argument takes an afternoon. Running it against your actual product is the part nobody does. Here's the audit, step by step, using the five commands in the Purple Cow skill for Claude Code (and other AI assistants). Budget about an hour; use your real product, not a hypothetical one.

Step 1: Audit what you actually have (/remarkable-audit)

Start with the honest baseline. The audit evaluates your product against Godin's criteria: Is it genuinely unusual in its category? Is it designed to be talked about? Does it serve one specific group extraordinarily well instead of many groups adequately? Is there a detail a customer would specifically mention in conversation?

Write down the answers before you run anything else. Most founders discover the same gap: the features they'd list in a pitch are the category's table stakes. The audit's job is to separate "good" from "remarkable" — they're different axes, and only one of them spreads.

Step 2: Find where safe is risky (/safe-is-risky)

Godin's central provocation is that the riskiest move in a crowded market is playing it safe. This step inverts your current strategy: where are you prioritizing short-term safety — matching competitors, sanding off the edges, avoiding the feature some customers might dislike — in ways that guarantee long-term invisibility?

This is usually the uncomfortable step, and it's the one that tells you whether you have a positioning problem or a courage problem. The output is a list of specific places where caution is costing you differentiation.

Step 3: Name your otaku (/otaku-finder)

Godin borrows "otaku" for the people obsessed with a category — the ones who seek out the best version of a thing and talk about it constantly. They matter more than your average buyer because they spread ideas for free.

The otaku finder builds that profile for your category: who the obsessed buyers are, what they value, where they gather, and what they argue about. If your answer to "who is this for?" is "small businesses" or "busy professionals," this step is where the audit pays for itself — sneezers can't spread a product aimed at everyone.

Step 4: Generate the edgy options (/edgy-brainstorm)

Now generate remarkability candidates: ten specific ways your product, service, or marketing could be extreme in some dimension — the version that makes someone stop and say "have you seen what they're doing?" Not ten incremental improvements. Ten things a competitor would call too much.

You won't ship most of them. You need the list because remarkable ideas rarely arrive one at a time; they show up when you force yourself past the safe five.

Step 5: Design the spread (/sneezer-strategy)

Finally, design for the people who'll carry it. Sneezers are the otaku who actively tell others. This step turns your chosen remarkable element into a word-of-mouth plan: the product moments, features, and incentives that make sharing easy and natural for that specific audience — not a generic "referral program," but sharing designed around what your otaku already talk about.

The sequence matters

Run them in order: audit → safe-is-risky → otaku → brainstorm → sneezer strategy. Skipping the audit produces brainstorms untethered from your actual weaknesses; skipping the otaku step produces ideas designed for an audience that doesn't exist.

Run it on your product

The Purple Cow skill runs this full audit as five guided commands in your AI assistant, using your answers as the input — it's $7 one-time in the BookSkills library. New to book skills? Start with the free Atomic Habits skill to see the format first. And if you end up running audits like this on a shelf of books, Annual Access unlocks all 43 skills for $79.99/year — it pays for itself at 12 single skills ($84 bought one at a time), so it's the right call only if you'll actually use a dozen or more.