An FAQ section helps depth when it answers real questions the body didn’t, and pads the page when it just restates the article in question form. That single line is the whole test, and the reason FAQ blocks have such a mixed reputation is that they get used both ways. Used well, an FAQ resolves the questions a reader still has after the main content: the edge cases, the “what if,” the related decision, the objection that didn’t fit the article’s through-line but that searchers genuinely ask. Each of those entries adds coverage the page didn’t have, and that is real depth.
Used as padding, an FAQ takes points the article already made and reformats them as questions. “What is X?” gets answered with a sentence the introduction already said, so the block adds words and question marks but no new information. To a reader it feels like filler, and to Google it adds nothing the page didn’t cover. Worse, when the same FAQ block is copied across many pages, the duplication can dilute those pages rather than strengthen any of them.
So the test is per entry, not per block. Read each question and ask whether its answer gives the reader something the body doesn’t already say. If it does, it belongs. If the answer just echoes the article, it is padding wearing a question mark, and a question the page already answered isn’t really a frequently asked question. A short FAQ of genuinely adjacent questions beats a long one that paraphrases the page back to itself.
Go through the FAQ one entry at a time, keep the questions that resolve something the body left open, and cut the ones that restate what you already wrote.