Should all my FAQs live on one page, or should each question get its own page?
Give each substantive question its own URL and group them into clusters. A single mega-FAQ dilutes every answer, forces retrieval to choose among twenty competing passages on one document, and leaves you with one citable URL for forty questions. Separate pages let each question own its title, heading, structured data and internal links. Keep a hub page listing them all - the hub earns navigation and crawl paths, the individual pages earn citations.
What goes wrong on the mega-FAQ
Three things. The page title can only describe one thing, so thirty-nine questions get a title that does not match them. Chunks from the same document compete with each other during re-ranking. And when an assistant does cite you, the link drops the reader at the top of a wall of unrelated questions.
There is also a practical editorial problem: nobody ever revises a question buried on line 340 of a page. Individual pages get maintained because they can be seen.
There is a measurement cost as well. With everything on one URL, you cannot tell which questions are earning attention and which are dead weight, because all the data collapses into one row. Separate pages give you per-question performance, which is what tells you where to write next.
Where a combined FAQ still earns its place
Short operational questions belong inline on the service page they relate to. Do you charge for estimates. What is your service area. Do you work weekends. These have no nuance and no follow-up.
A workable rule of thumb: if the honest, complete answer runs under about forty words and carries no tradeoff worth explaining, put it inline. If it needs three hundred words because the real answer is 'it depends, and here is what it depends on', it is a page.
The hub page is doing real work
A cluster hub gives crawlers a single dense path to every answer, gives readers a way to browse laterally, and lets you show related questions on each answer page. Those related links matter more than they look: they are how a reader who arrived from an assistant on one narrow question discovers that you have answered the next four.
That structure is the whole design of an answer hub, and it is why it behaves differently from a blog archive sorted by date.
Cluster structure also makes maintenance tractable. When a regulation changes or you add a service line, you can find and revise the six affected answers instead of scanning a monolith. Content that cannot be maintained stops being accurate, and inaccurate content is worse than none in a system that rewards consistency.
Do not create pages you cannot answer well
The failure mode of this approach is a hundred near-identical thin pages generated to fill a matrix. They compete with each other, they teach nothing, and they make the whole domain look mass-produced.
The test is simple and unforgiving: does this page contain at least one thing a knowledgeable reader did not already know? If not, fold it into a neighbor. Volume without substance is the fastest way to make automated content actively harmful.
Topics: site architecture · FAQ · content clusters · internal linking
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.