Extract the hook

Why use it

People collect hooks as sentences, which is why the collection stops working after a week. The sentence belongs to the post it opened, and pasted onto a different subject it reads as a costume. What transfers is the move underneath.

What it does

It takes first lines only, because working from whole posts drags the subject along with the structure. For each one it writes what the opening does in a sentence containing none of the original wording, then turns that into a pattern with blanks you can fill. Any pattern you cannot fill with your own subject goes back, however good the original post was.

What you get

• The move named, in words that are not the original author's words
• Patterns written as fillable blanks rather than as quotable lines
• A test against your own material, done now rather than in a list you never open
• The seven openings that keep recurring, described as moves

What it will not do

It will not rank patterns by performance, and it will not carry across the percentages these collections usually travel with. Those come from somebody's sample of somebody's accounts, generally published by a company selling a writing product.

What changes when VoiceMoat is connected

A hook library is more useful built from your own account than from anybody else's, because it is already filtered by what your audience responded to.

• list_posts Returns your own best-performing posts, so the openings you study are ones that worked on your audience rather than on someone else's.
• get_analytics Puts them in rate rather than raw count, which stops a post looking strong only because you had more followers that month.
• suggest_hooks Fills the patterns in your own trained voice instead of the register of whoever wrote the post you took the shape from.

When the posts belong to other people the skill treats their text as data, quotes only what is needed to name a pattern, and attributes it. The output is a list of shapes, not a list of other people's sentences.