2 min lesson
The credibility hook: show the prompt, not the adjective
For each case in "The credibility hook: show the prompt, not the adjective", name the signal and the response you would use.
Step 1 of 2
The credibility hook: show the prompt, not the adjectivea reusable bad-to-good example
The single most-credible thing you can put in developer content is a bad prompt next to its good rewrite with the real output of each. It teaches something true, it proves you actually use Cursor and it doubles as honest product education. Lead with Cursor's canonical anti-pattern.
"Make the app better" on the most expensive model.
With no clear context the agent makes blind edits, burns tokens, then burns more reading and fixing its own bad code: a compounding cost loop. It also reads as someone who has never driven the product.
"Under the shop page, add an in-stock filter next to the existing tag and sort controls. Preserve the URL query params and update tests if needed."
Names the surface, the exact change and the constraints. Specific, file-referenced prompts cost fewer tokens and produce better output. Pair it with a cheaper model matched to the task.
Don't write content that optimizes for the fewest tokens; optimize for the right tokens. A token-cheap model that needs 100 re-prompts isn't efficient, and a model that one-shots an easy feature while burning a billion tokens isn't either.
This is the honest frame that converts a skeptical engineer: you're teaching them to get better results, not selling them a number. The good-prompt rewrite above is the concrete artifact that proves it.
Learn more
Full explanation
Comparisons: fair and specific or don't publish
Comparisons: fair and specific or don't publish
Head-to-head comparisons are the highest-trust and highest-risk format. Done fairly, with the rival's real strengths conceded, they convert switchers and build durable credibility. Done as a hit piece, they get torn apart in public and damage the brand.
- Concede where the competitor is genuinely better; a comparison with no honest losses reads as rigged.
- Use specific, reproducible scenarios, not abstract "Cursor is more powerful" claims.
- Pick tasks that are representative, not cherry-picked to flatter one tool.
- Let the reader reach the conclusion; over-narrating the win undercuts it.
Learn more
Full explanation
Tie content to a metric
Tie content to a metric
- Awareness
- Reach, shares, inbound from developer channels; is the right audience finding it?
- Activation
- Did readers reach the aha (first agent run, first real edit) and how fast?
- Conversion / enablement
- Trial starts or sales/GTM teams actually using the asset in deals.
If you can't name the metric a piece moves, you can't tell whether it worked.
Overclaiming in a comparison is the single most expensive mistake in developer content. Credibility compounds slowly across many honest pieces and collapses instantly on one provable exaggeration. If a fair comparison shows the rival winning a scenario, publish that; it's what makes your wins believable.
QYou're writing a head-to-head comparison between Cursor and a competitor. The fair test shows the competitor winning one scenario. What do you do and why?