tagged PDF
A tagged PDF carries the structure of its own text beside the drawing: this is a heading, this is a table, this cell belongs to that column.
In more detail
Underneath, a PDF is a list of orders: put this word at this spot, draw this line there. It is a picture of a page, and a picture does not know what it shows.
The tags are a second list, running alongside the first. They say which words together make a title, which make a paragraph, where a table starts and where each of its cells belongs. The drawing does not change. What changes is that the file now knows what it drew.
What it does on a real document
On a table, the tags say which row holds the headings and which column each cell sits under. Software pulling the figures back out gets a table, instead of a scattering of numbers at coordinates.
They also fix the order to read the page in, which need not be the order it was drawn in. On a page in two columns the drawing runs down one column and then the other, or jumps between them; the tags say which of the two it is.
Where it comes from, and why
It came into the format in 2001. Two needs pushed it at the same time: reading a page aloud to somebody who cannot see it, and reflowing a page onto a small screen, which was a new problem then.
It is the ground the accessibility rule was later built on. PDF/UA does not invent the structure: it says which structure a document owes its reader, and how that structure has to be written.
See it on a finished document
A document that can be read aloud A list whose pictures speak too The page that goes further on this word