On this page
The groundwork for good answers
The search never reads whole documents. It reads chunks — the pieces your content is split into.
This split decides the quality of every answer. Pieces that are too big bury the answer in noise. Pieces that are too small lose their context. Wisplex splits your content automatically into pieces of the right size — you never have to think about it.
The origin stays attached
A piece of text is only useful if you know where it comes from. That's why every chunk keeps its origin:
- The source page for imported web content.
- The page number for PDFs.
- The collection it belongs to.
This is what lets every answer name its source — and a cited PDF open at the right page.
The full page when it matters
Splitting has one risk: the matching piece is not always the piece with the whole answer. Wisplex covers this. For imported web pages, a matching chunk is expanded to its full page during search. If a question matches the contact section of a page, the opening hours from the section next to it come along too.
Together with semantic search, this turns a pile of documents into answers that are both precise and complete.