Because they are built from a keyword list without checking what exists. A perfectly good brief gets written for a term the site already ranks for with a three-year-old post.
The new page splits the signal, both drop, and the honest conclusion — that the old one should have been updated — arrives a quarter later.
Queries are clustered by what actually ranks for them, and each cluster is mapped to the page that serves it. Before anything is commissioned, that mapping says whether a page already exists.
Gaps are then separated into missing pages and underperforming ones, which are different jobs with different costs.
It will not write the content, and the briefs are a specification rather than a draft.
And it will not settle an editorial argument. It supplies evidence about what is already there, which usually shortens it.
The backlog is ordered by expected value over effort with the reasoning recorded, so somebody can see why an item is first rather than being told.
And updating three existing pages can legitimately outrank writing a new one, which a keyword list alone will never suggest.
Yes, through keyword-to-page mapping, before anything is commissioned.
Update, more often than teams expect.
No. Briefs are a specification; the writing is yours.
We will run it end to end on your own numbers in half an hour, and tell you honestly which parts Treepie does not improve.