warpdotdev draft-article-title
Read the Google Doc, internalize the core argument, then generate 8–12 title candidates scored against the engagement data below. Include the user's initial idea labeled [Initial idea].
Fetch the doc with:
gws docs documents get --params '{"documentId": "<ID>"}' --format json
Doc ID is the string between /d/ and /edit in the URL.
Historical Engagement Data
Maintain a table of the author's past article titles and engagement so titles can be scored against what actually performed. Replace the synthetic rows below with your author's real data (keep it in a private location if it is confidential).
Engagement rate = (likes + replies + reposts + quotes + bookmarks) / impressions.
| Title | Impressions | ER | Bookmarks |
|---|---|---|---|
| Example: a contrarian, specific wedge title | 50,000 | 2.4% | 500 |
| Example: "what comes after X" framing | 30,000 | 2.6% | 440 |
| Example: operator decision framework | 30,000 | 3.2% | 670 |
| Example: vague philosophical claim (underperforms) | 10,000 | 0.9% | 25 |
Patterns to use: operator decision frameworks, specific technical problems, "what comes after X", contrarian + specific wedge, category-defining claims, bold imperatives.
Avoid: abstract business-model theory, generic pain without a fresh angle, vague philosophical claims.
Angles to cover
Generate at least one title per angle: direct statement · "what comes after X" · contrarian · founder/brand voice · operator framework · bold imperative · question form.
Output
Score each title 1–5 on Clarity, Reach, ER Potential, Fit. One-sentence note per row. Then recommend top 3 and explicitly place the initial idea.