Most people open Claude and ask it to write something. The practitioners getting the most out of it are using it to argue back at them.
A growing number of product builders, marketers, and freelancers are treating Claude less like a content machine and more like an on-call strategist. The pattern that keeps coming up: describe a feature you're about to build, and ask Claude why it might fail. Describe a pricing model, and ask it to find the edge cases where customers will feel cheated. Write three versions of a headline, and ask which one a skeptical reader would ignore first.
This isn't Claude doing the creative work. It's Claude doing the critical work.
Where This Shows Up in Practice
The UX strategy angle is particularly common. Instead of asking Claude to generate wireframe copy or user personas, practitioners are walking it through a product decision - why this feature, why now, who specifically is it for - and asking it to identify the gaps. Claude tends to ask clarifying questions, flag unstated assumptions, and surface the "but what about..." cases that get skipped in sprint planning.
Pricing model testing is another use case that comes up repeatedly. Describe your tiers, your usage limits, your upgrade triggers, and Claude will tell you where a cost-conscious customer is likely to feel burned. It doesn't replace a proper pricing study, but it catches the obvious traps before you publish a pricing page.
Copy variation feedback is more straightforward: write five subject lines, ask which two a tired person scrolling email would actually open and why. The reasoning Claude provides is more useful than a thumbs up or down - it tends to name the specific word or framing that weakens each option.
The Reason It Works Better Than Expected
Claude was trained on an enormous amount of text that includes business writing, product debates, design critiques, and strategic analysis. When you ask it to steelman a bad idea or find the weak point in an argument, it draws on that knowledge rather than just agreeing with you.
The practical upside: it is available at 2am before a product launch, has no stake in being liked, and will not soften criticism to preserve a relationship. That combination is more useful than it sounds when you are the only person reviewing your own work.
The limitation worth keeping in mind: Claude does not have access to your actual customer data, your analytics, or your competitive position. It can find logical gaps, but it cannot tell you whether the gap you found is the one that actually killed similar products. Pair it with real user feedback and it becomes a sharper tool. Use it alone and treat its concerns as hypotheses to test, not verdicts.