In this lesson you
- Crawled a website and listed every PDF on it
- Checked a representative sample of up to twelve in full
- Read where they stand, section by section of the site
Transcript
Your website
The free assessment. Before you can plan the work, you need two answers. How many PDFs are on your website? And how far are they from accessible? The free assessment answers both, in two stages. First it finds every PDF on your site. Then it checks a sample of them in full. It starts with your site's address, in Crawl a website. Enter the address. Proof follows your pages from there, the way a visitor clicking links would. First, confirm that you own the site, or that its owner has authorized you. Proof honors robots.txt, the file where a site says what crawlers may visit. Then choose Start crawl. By default, the crawl only discovers. It builds the list of PDFs and stops there, so nothing is downloaded until you've seen the list.
The inventory
When the crawl finishes, the top of the page says how much of the site it covered. Here, Crawl complete, whole site covered. Proof visited four pages and found fourteen PDFs on Maple Ridge's site. Each is listed with its name, address and size. This inventory counts every PDF, however many there are. To keep the list for your records, choose CSV. It downloads as a spreadsheet.
The sample
Checking documents in full takes longer than listing them. So the free assessment checks a sample, of up to twelve. Import assessment sample picks them for you. It spreads them across your site's sections, sizes and kinds of document, so they stand for the rest. On a small site like this one, the sample can be most of the site. On a large one, it's a cross-section.
Where they stand
Once the sample has been checked, the Corpus Assessment shows where those documents stand. The headline is the share that are already compliant with WCAG 2.1 AA, the accessibility standard. Here, two of the twelve. Beside it are those that failed, with a confirmed violation, and those that need a person to review them. The next question is what it would take to fix the rest. What stands open, and who owns it, answers that. Every open violation is counted once, by who moves it next. Some clear automatically, with no one's input. For some, AI can draft the fix, and a reviewer accepts it. The rest need a person to confirm them. Here, that's seven that Proof clears on its own, three where AI drafts, and four for a person. And one waits until another fix lands first. You'll also want to know where on your site the problems are. Exposure by site section shows that, for sections like the clinics pages. Documents in no section are grouped as Unfiled. For each section, it counts the documents that failed, need review, or are compliant. Auto clear counts documents the automatic pass would fix completely. That's the quickest work to plan. Last, so anyone can trust the numbers, the report records how the documents were gathered. It names the pages followed and the addresses found serving PDF files, and counts copies of the same file once. Under that, Scope of these checks says what the assessment covers and what it doesn't claim. It isn't a full conformance audit.
Taking it further
Now you know how many PDFs your site has, where they stand, and how much of the work Proof can do on its own. To read every part of this report in detail, or to share it with your team, see The corpus assessment and Share assessment.
Recap
That's the end of this lesson.
Maple Ridge Community Health, its documents and its people are fictional, made for these lessons.