There are many ways to define a workflow for PDF remediation. Every document is unique with unique contents and PDF elements. No step-by-step process universally applies to every document. In a nutshell, the document itself will define the optimal workflow.
There are, however, certain "orders of operations" that tend to be most efficient.
Looking at a new document, you should try to first understand how large it is and how consistently formatted the individual pages are (i.e. a newly created Word document from a clean template versus a 10-year-old Frankenstein document compiled from 5 different sources).
-
Evaluate existing tag structure
After determining a general sense of how large and complex the document is, make a decision about the existing tag structure (if applicable).
If the document was previously tagged, is that tag structure worth maintaining? -- or is it faster to just use the Zone Detection slider and let Equidox recreate the reading zones?
If the Zone Detector is used, typically it makes sense to apply the sensitivity level to all pages -- this will theoretically give a good, accurate zone layout on all of the subsequent pages (if there are outlier pages, the user can always revisit the Zone Detector to make an adjustment on each specific page).
-
Defining Headings
On the page level, the first thing to look for is headings. This allows the user to immediately develop some structure (this also helps when looking at the HTML preview because the large font headings are clear and obvious to help validate the reading order for the other elements on the page).
When defining headings, think back to the formatting/template of the document -- if heading styles are consistently used throughout the document, it makes sense to press the "use as template" checkbox on the H2s, H3s, H4s, etc.
Once the page is saved, Equidox will apply those heading levels to zones that have the same font characteristics on subsequent pages in the document.
-
Images
Next, it usually makes sense to focus on the images. If the images are decorative, hit the "Images" checkbox under the Zone Detection slider to artifact all of the images on the page with one click. Or, select each individual image zone and hit "Backspace" on the keyboard to artifact the single image. Then finish the images by adding alt-text wherever necessary.
-
Tables, Lists, and Footnotes
Next, address any tables or lists, or footnotes on the page (in whichever order is preferred).
-
Reading order should be last
Finally, check the reading order. The reason to wait until the end to adjust the reading order is that if / when a new reading zone is created, by default it will slot itself into the lowest available number. So if there were 20 zones automatically generated on a page, when a new zone is drawn, it will take on Zone 21 within the sequence.
The best workflow is to first capture all of the content inside of reading zones and make the final touchups to the reading order (if necessary) at the end. That way, it is only necessary to have to adjust the reading order once per page at a maximum.
When setting the reading order, first use the full-page re-order button to get the zones as close as possible. Then adjust the individual zones using the decimal system to target outlier zones that don't perfectly align with the layout of the rest of the page (i.e. an image in the center of the page that the end user might want to read at the end of the reading order).
If a page is a simple top-to-bottom / left-to-right layout, the reading order will not need to be touched at all -- Equidox will automatically set it by default when arriving at the page.
Every document is a puzzle to solve
Use the software for a couple of hours or remediate a few dozen pages -- and the steps that have been listed above become second nature. Part of what makes PDF accessibility challenging is that all PDFs are unique from one another. When looking at remediation like solving a digital puzzle, Equidox is a tool that will help solve the puzzle in a shorter amount of time.